INDEX

Explanations

phrases related to buildings and their physical features

oai_token-act-pair · gpt-3.5-turbo

references to multiple-story buildings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ateurs

-0.81

ugal

-0.80

ggies

-0.77

istered

-0.77

ournal

-0.74

uters

-0.74

estern

-0.71

ensable

-0.71

ocol

-0.69

eeks

-0.68

POSITIVE LOGITS

story

1.11

telling

1.02

LIN

0.91

line

0.88

LINE

0.80

lined

0.79

glass

0.79

 Stories

0.78

 Story

0.77

lines

0.76

Activations Density 0.009%

phrases related to buildings and their physical features

references to multiple-story buildings

No Comments

No Known Activations