INDEX

Explanations

phrases related to public opinion and political discourse

oai_token-act-pair · gpt-3.5-turbo

references to public opinion and voter sentiments

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tein

-0.81

HQ

-0.75

 retracted

-0.74

fell

-0.73

namese

-0.68

 Sisters

-0.67

 siblings

-0.64

 ministries

-0.63

 aliases

-0.63

 protocol

-0.62

POSITIVE LOGITS

 clam

0.97

 overwhelmingly

0.90

 electing

0.84

 Votes

0.84

 dissatisfied

0.83

 revolt

0.82

 vote

0.81

 votes

0.81

 Tuls

0.75

 voting

0.74

Activations Density 0.832%

phrases related to public opinion and political discourse

references to public opinion and voter sentiments

No Comments

No Known Activations