INDEX

Explanations

quantitative descriptions involving approximations

oai_token-act-pair · gpt-3.5-turbo

the word "about" and specify quantities or measures

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

è¦

-0.61

oli

-0.54

onsense

-0.54

 dstg

-0.52

BT

-0.52

Others

-0.50

OTUS

-0.50

elist

-0.50

ework

-0.49

apolis

-0.48

POSITIVE LOGITS

 halfway

1.02

0.91

 half

0.89

 midway

0.88

0.87

Activations Density 0.090%

quantitative descriptions involving approximations

the word "about" and specify quantities or measures

No Comments

No Known Activations