INDEX

Explanations

phrases related to going back in time or reversing a decision

oai_token-act-pair · gpt-3.5-turbo

references to going back, often in a nostalgic or reflective context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

anooga

-0.70

rez

-0.63

uracy

-0.60

Us

-0.58

Es

-0.57

Eh

-0.56

 Consent

-0.55

Neg

-0.54

kes

-0.54

orst

-0.54

POSITIVE LOGITS

stab

1.03

tracking

0.94

packs

0.93

packing

0.92

dated

0.90

track

0.86

doors

0.84

trace

0.84

wards

0.83

gam

0.82

Activations Density 0.023%

phrases related to going back in time or reversing a decision

references to going back, often in a nostalgic or reflective context

No Comments

No Known Activations