INDEX

Explanations

instances of the word 'round', possibly in the context of discussions, actions, or objects

oai_token-act-pair · gpt-3.5-turbo

references to organized discussions or meetings

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

 destro

-0.78

 Cald

-0.68

 indoctr

-0.64

ionage

-0.64

 Guilty

-0.63

WN

-0.60

 Norn

-0.59

ful

-0.59

earch

-0.59

 offended

-0.58

POSITIVE LOGITS

table

1.14

trip

1.11

abouts

1.09

about

1.07

outs

0.97

robe

0.87

meal

0.86

ups

0.86

ings

0.85

worms

0.83

Activations Density 0.015%

instances of the word 'round', possibly in the context of discussions, actions, or objects

references to organized discussions or meetings

No Comments

No Known Activations