INDEX

Explanations

specific nouns related to various activities or events

oai_token-act-pair · gpt-3.5-turbo

key terms related to legal actions and regulations

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

srf

-0.67

aturdays

-0.57

 alike

-0.56

Ĭ±

-0.56

ecause

-0.53

\\\\\\\\

-0.53

incinn

-0.49

regon

-0.49

ventus

-0.49

 Everywhere

-0.48

POSITIVE LOGITS

ieth

0.64

iest

0.60

osphere

0.56

liest

0.56

 itself

0.54

eus

0.54

washer

0.52

 Canaver

0.51

enge

0.51

's

0.51

Activations Density 0.827%

specific nouns related to various activities or events

key terms related to legal actions and regulations

No Comments

No Known Activations