INDEX

Explanations

phrases indicating a high degree of emphasis or importance

oai_token-act-pair · gpt-3.5-turbo

the expression of emphasis in statements

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

neapolis

-0.77

rones

-0.73

 GOODMAN

-0.71

venient

-0.67

æ°

-0.66

itures

-0.66

Many

-0.66

ERY

-0.65

OA

-0.65

ourse

-0.65

POSITIVE LOGITS

 appreciated

1.16

 alive

1.03

 intact

0.87

 alike

0.86

 reliant

0.85

 deserved

0.82

 dependent

0.81

 depended

0.80

 regarded

0.79

 resembled

0.79

Activations Density 0.046%

phrases indicating a high degree of emphasis or importance

the expression of emphasis in statements

No Comments

No Known Activations