INDEX

Explanations

phrases indicating advice or recommendation

oai_token-act-pair · gpt-3.5-turbo

the word "since" used frequently in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pta

-0.76

amina

-0.71

ereo

-0.70

BILITIES

-0.65

rawdownloadcloneembedreportprint

-0.64

Ruby

-0.64

ÙĴ

-0.64

©¶æ¥µ

-0.63

hack

-0.63

atives

-0.63

POSITIVE LOGITS

rely

1.32

ĸļ

0.90

userc

0.75

 sshd

0.72

0.71

pite

0.67

 mistakenly

0.63

 they

0.62

dfx

0.61

Activations Density 0.046%

phrases indicating advice or recommendation

the word "since" used frequently in various contexts

No Comments

No Known Activations