INDEX

Explanations

details related to specific actions and incidents

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

mentions of personal experiences and interactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

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

 approximation

-0.54

 Grail

-0.54

 newcomers

-0.54

othermal

-0.53

 adherents

-0.52

interstitial

-0.51

 accompan

-0.50

 nascent

-0.50

 currents

-0.49

 resumes

-0.49

POSITIVE LOGITS

 yesterday

1.58

 last

1.36

 earlier

1.24

ago

1.19

 LAST

1.16

terday

1.03

Yesterday

0.94

:(

0.94

last

0.92

 Yesterday

0.89

Activations Density 0.963%

details related to specific actions and incidents

mentions of personal experiences and interactions

No Comments

No Known Activations