INDEX

Explanations

references to media companies or entities

oai_token-act-pair · gpt-3.5-turbo

mentions of media-related entities and organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ãĤ¯

-0.75

ãĤ¡

-0.74

comings

-0.73

lier

-0.70

lift

-0.69

python

-0.68

sql

-0.68

liness

-0.68

ktop

-0.66

Synopsis

-0.66

POSITIVE LOGITS

eval

0.85

 outlets

0.84

 Arts

0.80

 Matters

0.78

 Mole

0.76

 Culture

0.75

Ltd

0.74

 Relations

0.74

 Noir

0.73

Buzz

0.71

Activations Density 0.015%

references to media companies or entities

mentions of media-related entities and organizations

No Comments

No Known Activations