INDEX

Explanations

text related to individuals identifying themselves in various ways, such as their political views or personal characteristics

oai_token-act-pair · gpt-3.5-turbo

phrases related to self-identification and political affiliations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

hiba

-0.75

 pretext

-0.70

MRI

-0.68

 temptation

-0.65

ousse

-0.63

 payoff

-0.62

pload

-0.62

 surfaces

-0.61

 surface

-0.61

ousand

-0.61

POSITIVE LOGITS

 adherent

0.89

odox

0.88

 progressive

0.85

 tolerant

0.84

 outsider

0.82

 fearless

0.82

 enlightened

0.81

 libertarian

0.81

 conscientious

0.79

 feminist

0.78

Activations Density 0.177%

text related to individuals identifying themselves in various ways, such as their political views or personal characteristics

phrases related to self-identification and political affiliations

No Comments

No Known Activations