INDEX

Explanations

names of individuals

oai_token-act-pair · gpt-3.5-turbo

references to individuals in a narrative context

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

undo

-0.72

AGE

-0.70

tnc

-0.64

reach

-0.64

definition

-0.63

NRS

-0.63

ulp

-0.62

IND

-0.60

 Meaning

-0.58

urg

-0.58

POSITIVE LOGITS

 oversaw

1.28

 oversees

1.23

 owns

1.23

 died

1.17

 presided

1.17

 famously

1.14

 preceded

1.14

 specializes

1.10

 testified

1.08

 served

1.07

Activations Density 0.094%

names of individuals

references to individuals in a narrative context

No Comments

No Known Activations