INDEX

Explanations

terms related to distance or level of advancement in relation to a certain standard

oai_token-act-pair · gpt-3.5-turbo

references to distance or extent

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

oxide

-0.74

ritten

-0.68

icer

-0.66

 TYPE

-0.64

ule

-0.63

 Outbreak

-0.61

uala

-0.60

icks

-0.60

onut

-0.59

itu

-0.59

POSITIVE LOGITS

 ahead

1.17

 apart

1.13

 away

1.11

 inland

1.10

fetched

1.08

 distances

1.05

away

1.01

forward

1.01

 Ahead

0.93

awa

0.92

Activations Density 0.070%

terms related to distance or level of advancement in relation to a certain standard

references to distance or extent

No Comments

No Known Activations