INDEX

Explanations

names and terms related to political figures and organizations

oai_token-act-pair · gpt-3.5-turbo

instances of the letter combinations indicative of linguistic variations or patterns

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

 Mehran

-0.83

hovah

-0.74

uni

-0.71

doi

-0.71

 Nare

-0.62

 reckoning

-0.61

hler

-0.61

utra

-0.60

 numbered

-0.60

éŃĶ

-0.58

POSITIVE LOGITS

enment

0.89

tarian

0.86

ateral

0.81

ibly

0.81

ional

0.78

ãĥ¼ãĥĨ

0.77

ables

0.75

ATION

0.72

 Stevenson

0.70

itely

0.67

Activations Density 0.235%

names and terms related to political figures and organizations

instances of the letter combinations indicative of linguistic variations or patterns

No Comments

No Known Activations