INDEX

Explanations

statements indicating the inability or refusal to provide a comment on a given topic

oai_token-act-pair · gpt-3.5-turbo

instances of the word "comment" and variations related to commenting on various topics

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

aez

-0.73

iHUD

-0.65

YP

-0.63

cot

-0.62

example

-0.61

fill

-0.61

adobe

-0.59

corn

-0.58

bey

-0.58

ãĥ¯

-0.57

POSITIVE LOGITS

 specifics

1.12

 publicly

0.98

 further

0.95

 afterward

0.86

 specifically

0.85

 extensively

0.85

 directly

0.84

 definitively

0.84

ispers

0.78

 regarding

0.77

Activations Density 0.035%

statements indicating the inability or refusal to provide a comment on a given topic

instances of the word "comment" and variations related to commenting on various topics

No Comments

No Known Activations