INDEX

Explanations

transitions indicating a conclusion or result

oai_token-act-pair · gpt-3.5-turbo

statements signaling conclusions or summaries

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

"},"

-0.62

 ______

-0.59

ga

-0.58

duction

-0.57

 Rumble

-0.56

 upgr

-0.56

GER

-0.55

 Barb

-0.55

 exclusively

-0.55

 burg

-0.55

POSITIVE LOGITS

fter

0.83

forth

0.82

pite

0.77

ername

0.76

forward

0.76

aucuses

0.74

ccording

0.73

entimes

0.72

hester

0.72

noon

0.70

Activations Density 0.034%

transitions indicating a conclusion or result

statements signaling conclusions or summaries

No Comments

No Known Activations