INDEX

Explanations

mentions of the word "column"

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

schild

-0.81

 Revenge

-0.68

acht

-0.63

gerald

-0.60

pict

-0.60

igree

-0.59

anian

-0.58

ickle

-0.57

iott

-0.57

guard

-0.56

POSITIVE LOGITS

§

0.76

ornia

0.71

ĳå£«

0.71

Ķ

0.71

BUS

0.69

rity

0.68

Ń·

0.68

ĺ

0.68

adelphia

0.67

cision

0.64

Activations Density 0.066%

mentions of the word "column"

No Comments

No Known Activations