INDEX

Explanations

mentions of the specific word "rows."

oai_token-act-pair · gpt-3.5-turbo

mentions of 'rows' or related contexts

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

 PLAN

-0.69

 circumstance

-0.68

natureconservancy

-0.64

 incentive

-0.64

 Scandinavian

-0.63

 Kenyan

-0.61

piece

-0.60

 retake

-0.60

 disparate

-0.60

 miracle

-0.59

POSITIVE LOGITS

rows

1.29

restling

1.04

icz

0.97

olver

0.93

ski

0.91

hip

0.90

lings

0.90

ight

0.88

rowing

0.88

abies

0.87

Activations Density 0.004%

mentions of the specific word "rows."

mentions of 'rows' or related contexts

No Comments

No Known Activations