INDEX

Explanations

words related to negative situations or states, such as crisis, distress, and repair

oai_token-act-pair · gpt-3.5-turbo

terms related to economic and social crises

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

 corrid

-0.66

 accur

-0.61

nice

-0.61

arlane

-0.60

amaru

-0.59

 diligent

-0.56

tein

-0.56

 Hemp

-0.56

nesday

-0.56

adden

-0.56

POSITIVE LOGITS

 financially

1.03

 owing

0.95

due

0.86

 footing

0.81

 economically

0.78

 Unable

0.77

 lately

0.75

 thanks

0.75

Recent

0.74

 amid

0.73

Activations Density 0.234%

words related to negative situations or states, such as crisis, distress, and repair

terms related to economic and social crises

No Comments

No Known Activations