INDEX

Explanations

phrases related to technical processes or instructions

oai_token-act-pair · gpt-3.5-turbo

words related to various trade and economic terminology

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

 Archdemon

-0.74

 Valkyrie

-0.68

 renown

-0.66

 reckoning

-0.65

 commem

-0.60

 anth

-0.58

 allowance

-0.58

 sciences

-0.57

 solvent

-0.56

 aware

-0.56

POSITIVE LOGITS

ulence

0.88

actionDate

0.87

acers

0.81

aughs

0.77

ctr

0.76

iste

0.76

Ped

0.73

acer

0.73

itars

0.70

dfx

0.70

Activations Density 0.110%

phrases related to technical processes or instructions

words related to various trade and economic terminology

No Comments

No Known Activations