INDEX

Explanations

phrases related to a comprehensive entity or collection

oai_token-act-pair · gpt-3.5-turbo

phrases related to different types of collections or aggregates

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

 Pradesh

-0.81

bia

-0.69

gyn

-0.67

uni

-0.62

 Savings

-0.62

vale

-0.61

ilet

-0.60

zai

-0.59

fan

-0.59

lez

-0.58

POSITIVE LOGITS

 medicine

0.79

arted

0.71

 knowledge

0.66

rency

0.63

milo

0.62

 evidence

0.61

 cards

0.59

 referen

0.57

 tailored

0.57

 questioning

0.56

Activations Density 0.094%

phrases related to a comprehensive entity or collection

phrases related to different types of collections or aggregates

No Comments

No Known Activations