INDEX

Explanations

mentions of the word "ro" specifically

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "ro"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Sons

-0.69

 Coulter

-0.68

ividual

-0.68

arians

-0.61

 eminent

-0.58

 Catalyst

-0.57

 EntityItem

-0.57

ivities

-0.57

 innocence

-0.56

DPR

-0.56

POSITIVE LOGITS

bably

1.20

spective

1.17

cks

1.13

spection

1.09

tto

1.05

vert

1.05

dden

1.04

blems

1.02

active

1.01

ject

1.00

Activations Density 0.033%

mentions of the word "ro" specifically

occurrences of the substring "ro"

No Comments

No Known Activations