INDEX

Explanations

names of specific entities or characters

oai_token-act-pair · gpt-3.5-turbo

references to specific names or entities, particularly related to sports

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Miko

-0.78

 Carbuncle

-0.70

 linem

-0.64

 Cuomo

-0.63

 Chimera

-0.62

 Hercules

-0.61

OTAL

-0.61

 Rebels

-0.61

 Â¶

-0.59

aling

-0.58

POSITIVE LOGITS

wyn

1.01

stad

1.00

kens

0.99

mar

0.96

sted

0.96

ham

0.93

ff

0.93

wich

0.92

haus

0.91

chel

0.91

Activations Density 0.057%

names of specific entities or characters

references to specific names or entities, particularly related to sports

No Comments

No Known Activations