INDEX

Explanations

terms related to inquiries or questions being made by people

oai_token-act-pair · gpt-3.5-turbo

instances of public reactions and responses to events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

thood

-0.72

 presided

-0.69

senal

-0.68

acl

-0.67

vati

-0.67

affer

-0.64

stra

-0.64

 overseen

-0.64

ainment

-0.63

=~=~

-0.62

POSITIVE LOGITS

 themselves

0.90

 louder

0.90

 clam

0.88

 selfies

0.80

 hashtag

0.80

 boycot

0.79

 chants

0.74

 boycott

0.74

 uproar

0.74

 hasht

0.74

Activations Density 0.474%

terms related to inquiries or questions being made by people

instances of public reactions and responses to events

No Comments

No Known Activations