INDEX

Explanations

mentions of giving paid speeches

oai_token-act-pair · gpt-3.5-turbo

mentions of speeches, particularly paid speeches

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 saline

-0.69

appropriate

-0.65

abol

-0.63

 Nation

-0.63

Captain

-0.62

 Mississ

-0.61

 Rutherford

-0.61

VAL

-0.61

 Wonderland

-0.60

astern

-0.59

POSITIVE LOGITS

 speeches

1.59

 lectures

0.96

 presentations

0.90

 speech

0.89

 concerts

0.88

 pitches

0.88

querque

0.87

anooga

0.85

gements

0.84

 seminars

0.83

Activations Density 0.012%

mentions of giving paid speeches

mentions of speeches, particularly paid speeches

No Comments

No Known Activations