INDEX

Explanations

mentions of the place "Kashmir"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "mir" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

YC

-0.73

 Lean

-0.68

ANY

-0.63

 Diversity

-0.63

gger

-0.63

0000000000000000

-0.62

 Tone

-0.62

growth

-0.62

Ô

-0.61

Camp

-0.60

POSITIVE LOGITS

ror

1.27

rors

1.05

acle

0.99

rored

0.98

acles

0.85

bled

0.82

ific

0.79

andum

0.78

abilia

0.78

abella

0.77

Activations Density 0.007%

mentions of the place "Kashmir"

instances of the word "mir" in various contexts

No Comments

No Known Activations