INDEX

Explanations

terms or phrases related to the concept of "returning."

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "return."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

hemat

-0.81

Cola

-0.78

fusc

-0.70

vern

-0.69

 Supplementary

-0.66

ciating

-0.66

aughtered

-0.66

Lab

-0.62

Ĭ±

-0.62

ascript

-0.62

POSITIVE LOGITS

return

0.96

Return

0.92

 return

0.92

 returns

0.87

ees

0.81

 Returning

0.79

 Return

0.77

backs

0.75

Returns

0.74

home

0.74

Activations Density 0.020%

terms or phrases related to the concept of "returning."

occurrences of the word "return."

No Comments

No Known Activations