INDEX

Explanations

strong emotional reactions or evaluations

oai_token-act-pair · gpt-3.5-turbo

expressions of emotional impact or reaction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

heed

-0.78

arta

-0.71

 defended

-0.65

HIP

-0.65

yt

-0.63

omorph

-0.62

 arsen

-0.61

 inhabit

-0.60

vati

-0.59

isa

-0.59

POSITIVE LOGITS

reddits

0.80

 tremend

0.79

nesday

0.76

éĹ

0.75

 abruptly

0.69

 initially

0.68

netflix

0.68

Wem

0.66

âĹ¼

0.65

 teasp

0.65

Activations Density 0.455%

strong emotional reactions or evaluations

expressions of emotional impact or reaction

No Comments

No Known Activations