INDEX

Explanations

words related to outcomes or results

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate outcomes or consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

tsky

-0.75

 Neigh

-0.56

 craw

-0.54

 neighbor

-0.54

 Neighbor

-0.53

crow

-0.53

 collaborator

-0.52

 edged

-0.51

pe

-0.51

ses

-0.51

POSITIVE LOGITS

antly

0.87

aneously

0.80

inally

0.80

ively

0.71

âĹ¼

0.70

uced

0.68

inate

0.66

in

0.66

 primarily

0.65

inations

0.65

Activations Density 0.038%

words related to outcomes or results

phrases that indicate outcomes or consequences

No Comments

No Known Activations