INDEX

Explanations

instances of the word "no" in different languages

oai_token-act-pair · gpt-3.5-turbo

negations or phrases that indicate absence or denial

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

Magic

-0.64

-$

-0.64

 wallets

-0.64

RAW

-0.62

 milliseconds

-0.61

 Fields

-0.59

Indiana

-0.57

ARY

-0.56

 Wizards

-0.56

 personalities

-0.56

POSITIVE LOGITS

lez

1.07

vez

1.06

nda

0.95

esi

0.92

tu

0.90

ovi

0.87

tz

0.85

vae

0.85

ku

0.83

 exting

0.83

Activations Density 0.044%

instances of the word "no" in different languages

negations or phrases that indicate absence or denial

No Comments

No Known Activations