INDEX

Explanations

questions related to specific topics or actions

oai_token-act-pair · gpt-3.5-turbo

questions and statements seeking clarification or confirmation

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

owners

-0.71

eka

-0.69

pora

-0.67

jay

-0.65

ama

-0.65

odcast

-0.64

elist

-0.63

uce

-0.63

heast

-0.62

reciation

-0.62

POSITIVE LOGITS

ãĤ»

0.78

irements

0.75

£ı

0.73

THING

0.72

Required

0.70

èª

0.68

ACTION

0.68

ãĥĬ

0.67

utable

0.67

Ïī

0.65

Activations Density 0.089%

questions related to specific topics or actions

questions and statements seeking clarification or confirmation

No Comments

No Known Activations