INDEX

Explanations

instances where a question or situation is introduced with the word "When" followed by an action or scenario

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "When."

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

agin

-0.68

number

-0.64

hash

-0.62

alle

-0.62

harm

-0.62

rolet

-0.61

ymph

-0.59

nut

-0.58

ãģ®é

-0.58

sonian

-0.57

POSITIVE LOGITS

soever

1.52

irlf

0.91

quickShipAvailable

0.81

bestos

0.81

 confronted

0.79

 asked

0.79

phalt

0.79

entimes

0.78

 simultane

0.76

IPS

0.75

Activations Density 0.037%

instances where a question or situation is introduced with the word "When" followed by an action or scenario

occurrences of the word "When."

No Comments

No Known Activations