INDEX

Explanations

specific locations or places

oai_token-act-pair · gpt-3.5-turbo

instances of the word "where."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ÂłÂłÂłÂłÂłÂłÂłÂł

-0.73

 squeeze

-0.69

ets

-0.63

 striking

-0.62

 punch

-0.61

 chronological

-0.61

ying

-0.59

 guessing

-0.58

yi

-0.57

astic

-0.56

POSITIVE LOGITS

upon

1.31

abouts

1.20

soever

1.04

fore

0.97

abama

0.79

izen

0.79

[|

0.78

ãĤ¼ãĤ¦ãĤ¹

0.73

Īè

0.72

VERTIS

0.70

Activations Density 0.066%

specific locations or places

instances of the word "where."

No Comments

No Known Activations