INDEX

Explanations

comparative phrases indicating a level or amount of something

oai_token-act-pair · gpt-3.5-turbo

references to levels or degrees of impact or effect

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

sbm

-0.74

tty

-0.69

ashed

-0.67

 Shuttle

-0.65

RAW

-0.61

blade

-0.60

 Ribbon

-0.57

 Grade

-0.57

 Riders

-0.57

imen

-0.57

POSITIVE LOGITS

mma

0.92

 extent

0.73

forth

0.70

iece

0.69

querque

0.67

ctuary

0.66

conom

0.65

 approximation

0.64

 contrad

0.64

fg

0.63

Activations Density 0.030%

comparative phrases indicating a level or amount of something

references to levels or degrees of impact or effect

No Comments

No Known Activations