INDEX

Explanations

words related to clothing items, especially shirts

oai_token-act-pair · gpt-3.5-turbo

words associated with police, limousines, and medical references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

PORT

-0.79

hyde

-0.68

 Dear

-0.65

 Patent

-0.64

 Spear

-0.63

 Walker

-0.61

 Falls

-0.61

FIELD

-0.61

 Spur

-0.61

 Hendricks

-0.61

POSITIVE LOGITS

inations

1.10

itational

1.02

ices

0.99

itating

0.96

utes

0.96

atorial

0.95

iency

0.94

ailable

0.94

icy

0.93

irtual

0.92

Activations Density 0.128%

words related to clothing items, especially shirts

words associated with police, limousines, and medical references

No Comments

No Known Activations