INDEX

Explanations

something with multiple characters, usually associated with a technical or specific context

oai_token-act-pair · gpt-3.5-turbo

the word "which" as a connector in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 VIDEOS

-0.73

MENTS

-0.67

Bas

-0.66

STE

-0.65

athi

-0.63

Behind

-0.62

politics

-0.58

GY

-0.57

BOX

-0.57

 Penalty

-0.55

POSITIVE LOGITS

 incidentally

1.08

 translates

1.06

 consists

1.06

 comprises

1.03

 specializes

1.03

 operates

0.99

 contains

0.97

 originated

0.96

 provides

0.95

 brings

0.94

Activations Density 0.125%

something with multiple characters, usually associated with a technical or specific context

the word "which" as a connector in various contexts

No Comments

No Known Activations