INDEX

Explanations

people and their roles within academic and research institutions

oai_token-act-pair · gpt-3.5-turbo

mentions of individuals involved in research or academic studies

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 decency

-0.74

theless

-0.74

 whine

-0.73

 fade

-0.66

 mirac

-0.66

 stale

-0.65

 haircut

-0.65

 intolerable

-0.62

 filthy

-0.62

 magically

-0.61

POSITIVE LOGITS

study

1.03

 study

0.95

 research

0.94

 studies

0.93

Study

0.91

doctoral

0.89

 resear

0.87

authors

0.86

research

0.84

earcher

0.81

Activations Density 0.209%

people and their roles within academic and research institutions

mentions of individuals involved in research or academic studies

No Comments

No Known Activations