INDEX

Explanations

the word stem "vers" appearing in different contexts

oai_token-act-pair · gpt-3.5-turbo

terms related to versions or variants

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

CHR

-0.65

 Pearce

-0.65

opathic

-0.64

redo

-0.63

Fra

-0.62

 judges

-0.62

urt

-0.61

UGH

-0.61

AFTA

-0.60

ļé

-0.60

POSITIVE LOGITS

atile

1.26

atility

1.19

erver

1.06

hip

1.06

terday

1.02

ational

0.94

pect

0.91

sein

0.90

chool

0.90

omething

0.89

Activations Density 0.024%

the word stem "vers" appearing in different contexts

terms related to versions or variants

No Comments

No Known Activations