INDEX

Explanations

news headlines or reports highlighting significant events

oai_token-act-pair · gpt-3.5-turbo

instances of the word "what" in various contexts

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

 largeDownload

-0.66

enance

-0.62

 Yards

-0.62

roads

-0.60

 Emirates

-0.58

 Gamb

-0.57

ohyd

-0.57

jee

-0.55

robe

-0.54

strap

-0.54

POSITIVE LOGITS

soever

1.06

 happened

0.99

 happens

0.96

 transpired

0.92

 constitutes

0.84

abouts

0.78

 kinds

0.77

 constituted

0.77

 happ

0.76

orate

0.74

Activations Density 0.094%

news headlines or reports highlighting significant events

instances of the word "what" in various contexts

No Comments

No Known Activations