INDEX

Explanations

references to clothing items, particularly shirts

oai_token-act-pair · gpt-3.5-turbo

references to shirts and shirt-related items

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

sys

-0.71

ths

-0.70

SPONSORED

-0.67

ruciating

-0.65

 ingred

-0.63

 Galile

-0.62

 fert

-0.62

ydia

-0.61

ITNESS

-0.61

OTOS

-0.60

POSITIVE LOGITS

leeve

1.20

hirt

1.20

 shirts

1.15

 sleeve

1.13

 sleeves

1.13

shirt

1.09

 shirt

1.07

 worn

1.02

shirts

1.01

 collar

0.97

Activations Density 0.044%

references to clothing items, particularly shirts

references to shirts and shirt-related items

No Comments

No Known Activations