INDEX

Explanations

references to agreements or widespread acceptance of an idea

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "consensus"

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

gin

-0.86

asus

-0.81

gun

-0.79

hib

-0.79

strip

-0.76

zona

-0.73

nov

-0.73

gins

-0.73

jiang

-0.71

asca

-0.70

POSITIVE LOGITS

 consensus

0.93

 unanimously

0.89

 amongst

0.82

 opinion

0.77

 among

0.77

ensical

0.75

ensus

0.72

ibel

0.71

 agreement

0.70

itarian

0.69

Activations Density 0.027%

references to agreements or widespread acceptance of an idea

occurrences of the word "consensus"

No Comments

No Known Activations