INDEX

Explanations

keywords related to specific people or entities called "Ag"

oai_token-act-pair · gpt-3.5-turbo

repeated occurrences of the token representing a specific category or type, likely indicating instances or mentions of entities or classifications

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Dare

-0.78

 Jenner

-0.75

 Shroud

-0.64

 jumper

-0.63

 sight

-0.63

 Cors

-0.61

 Tune

-0.60

éĹĺ

-0.60

 directions

-0.60

strap

-0.59

POSITIVE LOGITS

reement

1.41

reements

1.30

reed

1.21

encies

1.13

raham

1.07

nostic

1.05

ree

1.04

ressive

1.02

atha

1.01

rees

1.00

Activations Density 0.042%

keywords related to specific people or entities called "Ag"

repeated occurrences of the token representing a specific category or type, likely indicating instances or mentions of entities or classifications

No Comments

No Known Activations