INDEX

Explanations

instances of the word 'end'

oai_token-act-pair · gpt-3.5-turbo

terms related to repetition or recurrence

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

 pursu

-0.52

 Mehran

-0.48

imately

-0.48

 Lama

-0.46

 Mecca

-0.46

 lever

-0.46

mas

-0.46

gered

-0.46

 Omaha

-0.45

 Bentley

-0.45

POSITIVE LOGITS

OSP

0.64

lement

0.60

natureconservancy

0.59

apping

0.55

hig

0.55

bs

0.54

othe

0.54

abo

0.53

otin

0.53

unknown

0.53

Activations Density 1.006%

instances of the word 'end'

terms related to repetition or recurrence

No Comments

No Known Activations