INDEX

Explanations

phrases indicating a choice or decision point

oai_token-act-pair · gpt-3.5-turbo

phrases indicating choice or opinion, specifically contrasting options

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

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

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ourses

-0.66

inational

-0.66

xus

-0.66

vier

-0.64

marked

-0.62

urate

-0.61

umper

-0.61

runner

-0.60

ipal

-0.60

uph

-0.59

POSITIVE LOGITS

lando

0.88

acle

0.81

 hate

0.75

acles

0.74

Else

0.73

 Bust

0.73

not

0.72

hate

0.70

 starve

0.70

 lose

0.69

Activations Density 0.052%

phrases indicating a choice or decision point

phrases indicating choice or opinion, specifically contrasting options

No Comments

No Known Activations