INDEX

Explanations

references to celebrities or public figures and their activities or achievements

oai_token-act-pair · gpt-3.5-turbo

punctuations, specifically commas

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

VERTISEMENT

-0.81

Ł

-0.68

Ķ

-0.66

ãĤ´ãĥ³

-0.65

)}

-0.63

ITH

-0.62

},"

-0.62

":"/

-0.61

 msec

-0.61

RP

-0.60

POSITIVE LOGITS

Osc

0.67

 pandemonium

0.62

 Leilan

0.61

 Paste

0.60

dos

0.60

 Salvador

0.60

 Wilmington

0.60

 Verd

0.59

aka

0.59

 however

0.59

Activations Density 0.374%

references to celebrities or public figures and their activities or achievements

punctuations, specifically commas

No Comments

No Known Activations