INDEX

Explanations

keywords related to specific types or instances of objects or concepts

oai_token-act-pair · gpt-3.5-turbo

technical terms and concepts related to data representation and analysis

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

Negative Logits

å§«

-0.78

POSE

-0.67

ophers

-0.66

éĹĺ

-0.64

SOURCE

-0.63

 FANTASY

-0.61

adin

-0.59

é¾įå

-0.58

yss

-0.57

iasm

-0.55

POSITIVE LOGITS

matically

0.84

lessly

0.76

fully

0.67

lete

0.65

 akin

0.64

naire

0.63

ifies

0.63

("

0.62

boards

0.59

ously

0.58

Activations Density 0.747%

keywords related to specific types or instances of objects or concepts

technical terms and concepts related to data representation and analysis

No Known Activations