INDEX

Explanations

mentions of people or organizations, potentially in a negative context

oai_token-act-pair · gpt-3.5-turbo

the symbol or character representation of certain expressions or emphasis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 imitation

-0.89

 Seym

-0.72

 indo

-0.69

anium

-0.69

 mathemat

-0.68

 accompan

-0.66

 constitu

-0.66

 fortun

-0.66

arios

-0.65

 disadvant

-0.64

POSITIVE LOGITS

ï¸ı

1.23

ï¸

0.94

VER

0.85

å¥³

0.85

Balt

0.82

STEM

0.80

sure

0.76

legal

0.75

own

0.74

Â£

0.73

Activations Density 0.509%

mentions of people or organizations, potentially in a negative context

the symbol or character representation of certain expressions or emphasis

No Comments

No Known Activations