INDEX

Explanations

names of specific locations and groups related to political conflict and events

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names of places and significant 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Nguyen

-0.84

 Xander

-0.82

trace

-0.77

 Xperia

-0.76

 Wilde

-0.75

erest

-0.71

 ster

-0.70

TPP

-0.69

 Luthor

-0.68

Seattle

-0.67

POSITIVE LOGITS

iyah

1.53

 Mosque

1.38

 mosque

1.29

Ø©

1.25

ayn

1.14

yah

1.14

ibn

1.13

uddin

1.08

iya

1.08

Ibn

1.06

Activations Density 0.203%

names of specific locations and groups related to political conflict and events

proper nouns, particularly names of places and significant locations

No Comments

No Known Activations