INDEX

Explanations

a mixture of words and numbers with various characters and formatting that doesn't seem to have a specific pattern

oai_token-act-pair · gpt-3.5-turbo

non-standard characters, symbols, and email-related content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.68

ufact

-0.63

ģĸ

-0.63

ISA

-0.59

kus

-0.59

ãĤ¨ãĥ«

-0.58

haar

-0.57

elta

-0.56

 practition

-0.55

inctions

-0.55

POSITIVE LOGITS

enged

0.56

 bridges

0.54

 mirrors

0.52

iland

0.51

 stalls

0.50

 intermitt

0.49

clock

0.48

 nails

0.48

rollers

0.48

steamapps

0.47

Activations Density 1.343%

a mixture of words and numbers with various characters and formatting that doesn't seem to have a specific pattern

non-standard characters, symbols, and email-related content

No Comments

No Known Activations