INDEX

Explanations

words relating to directories, files, and technical specifications

oai_token-act-pair · gpt-3.5-turbo

repeated references to ordinal numbers and rankings

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

 Speedway

-0.74

staking

-0.70

 Petraeus

-0.65

cule

-0.65

 Faul

-0.62

gdala

-0.61

 Katrina

-0.61

 Canaver

-0.60

rition

-0.60

ropolis

-0.58

POSITIVE LOGITS

rd

1.03

irty

0.95

isl

0.92

yss

0.91

 baseman

0.86

oÄŁan

0.83

ouble

0.83

aughter

0.79

ividual

0.79

eal

0.77

Activations Density 0.008%

words relating to directories, files, and technical specifications

repeated references to ordinal numbers and rankings

No Comments

No Known Activations