INDEX

Explanations

locations or specific entities within the context of a larger entity

oai_token-act-pair · gpt-3.5-turbo

references to organizations, teams, or collective entities

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

 Helpful

-0.87

ricular

-0.85

pmwiki

-0.79

å§«

-0.69

iminary

-0.65

worldly

-0.65

actionDate

-0.64

ega

-0.64

utory

-0.63

 derog

-0.62

POSITIVE LOGITS

 itself

0.94

 catalogue

0.80

itol

0.79

ariat

0.78

 wagon

0.73

osphere

0.72

 empire

0.70

 logo

0.70

 pictured

0.69

 debacle

0.68

Activations Density 0.393%

locations or specific entities within the context of a larger entity

references to organizations, teams, or collective entities

No Comments

No Known Activations