INDEX

Explanations

words related to conflict, problems, or issues

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "trouble" and its variations in the context of negative situations or conflicts

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

oliberal

-0.79

oir

-0.72

urse

-0.71

ithe

-0.70

ulture

-0.70

orney

-0.68

entric

-0.66

emouth

-0.66

ivist

-0.66

icrobial

-0.65

POSITIVE LOGITS

hooting

1.13

makers

0.97

 troubles

0.90

maker

0.83

some

0.83

 trouble

0.80

fully

0.78

engers

0.75

flake

0.75

making

0.74

Activations Density 0.027%

words related to conflict, problems, or issues

occurrences of the word "trouble" and its variations in the context of negative situations or conflicts

No Comments

No Known Activations