INDEX

Explanations

expressions of gratitude

oai_token-act-pair · gpt-3.5-turbo

expressions of gratitude

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

FIELD

-0.70

Fa

-0.65

 Gong

-0.64

 booted

-0.64

 deed

-0.64

 Pagan

-0.62

 Rahman

-0.62

TY

-0.60

 broom

-0.60

 conformity

-0.59

POSITIVE LOGITS

awaru

0.77

isu

0.77

agu

0.76

romy

0.75

onom

0.73

ani

0.72

ql

0.68

olid

0.68

amy

0.68

imi

0.67

Activations Density 0.000%

expressions of gratitude

expressions of gratitude

No Comments

No Known Activations