INDEX

Explanations

values being assigned or manipulated in computer programming code

oai_token-act-pair · gpt-3.5-turbo

references to values and variables in programming

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

ocamp

-0.76

kefeller

-0.74

ooks

-0.74

dar

-0.73

nee

-0.71

osaurus

-0.71

spread

-0.69

ned

-0.68

riad

-0.68

asio

-0.66

POSITIVE LOGITS

 values

0.84

 value

0.84

EStream

0.73

values

0.71

 Value

0.69

 Values

0.66

VALUE

0.65

value

0.65

 output

0.65

 outputs

0.65

Activations Density 0.015%

values being assigned or manipulated in computer programming code

references to values and variables in programming

No Comments

No Known Activations