INDEX

Explanations

expressions or mentions of 'hotspots'

oai_token-act-pair · gpt-3.5-turbo

references to hot spots or highly active locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Rite

-0.67

 shapes

-0.66

©¶æ

-0.65

ulic

-0.65

ically

-0.65

ned

-0.65

rily

-0.63

 telev

-0.61

ruary

-0.61

ience

-0.60

POSITIVE LOGITS

pots

1.32

pot

1.24

pace

1.07

paces

1.05

peak

1.04

creen

1.01

pur

1.00

pad

0.91

hirt

0.88

apons

0.86

Activations Density 0.090%

expressions or mentions of 'hotspots'

references to hot spots or highly active locations

No Comments

No Known Activations