INDEX

Explanations

words related to advocating for or supporting a particular issue or belief

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "cause."

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

PDATE

-0.79

Ku

-0.72

 Seym

-0.68

illet

-0.67

egu

-0.66

 Seasons

-0.65

aeper

-0.65

awatts

-0.64

raph

-0.63

 Leopard

-0.63

POSITIVE LOGITS

 cele

1.37

cause

0.86

way

0.79

ality

0.79

 celeb

0.76

vier

0.72

facts

0.71

fare

0.70

forge

0.70

wagon

0.69

Activations Density 0.028%

words related to advocating for or supporting a particular issue or belief

references to the concept of "cause."

No Comments

No Known Activations