INDEX

Explanations

references to voting behavior and political positions

oai_token-act-pair · gpt-3.5-turbo

terms related to voting, polls, and decisions

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

ngth

-0.84

uctive

-0.76

¿½

-0.73

hai

-0.72

catentry

-0.72

attribute

-0.71

 Divinity

-0.69

OIL

-0.68

cci

-0.67

ãĤ¼ãĤ¦ãĤ¹

-0.67

POSITIVE LOGITS

 unanimously

1.01

 votes

0.93

 referendum

0.88

 vote

0.86

 ballot

0.85

 approving

0.84

 ballots

0.83

votes

0.82

 elections

0.81

 refere

0.80

Activations Density 0.170%

references to voting behavior and political positions

terms related to voting, polls, and decisions

No Comments

No Known Activations