INDEX

Explanations

specific names of entities

oai_token-act-pair · gpt-3.5-turbo

specific names or titles associated with entertainment and media

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

theless

-0.85

 referen

-0.76

 concerned

-0.74

 subscribed

-0.73

 conclud

-0.70

âĶĢâĶĢ

-0.69

 describ

-0.68

 logically

-0.67

 endowed

-0.66

 revolving

-0.66

POSITIVE LOGITS

zes

0.94

itars

0.92

eworks

0.88

indal

0.86

naire

0.85

omaly

0.84

itaire

0.83

 Designs

0.82

enstein

0.82

plates

0.82

Activations Density 0.370%

specific names of entities

specific names or titles associated with entertainment and media

No Comments

No Known Activations