INDEX

Explanations

phrases related to iterative processes or versions

oai_token-act-pair · gpt-3.5-turbo

references to iterations and iterative processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

FACE

-0.86

light

-0.79

lighting

-0.78

lights

-0.73

friends

-0.70

 Aram

-0.67

ening

-0.64

phone

-0.64

 jelly

-0.64

lord

-0.64

POSITIVE LOGITS

 iter

1.03

 Iter

0.86

 iteration

0.85

 iterations

0.79

atively

0.78

uably

0.78

issance

0.76

ival

0.76

Iterator

0.70

ively

0.70

Activations Density 0.022%

phrases related to iterative processes or versions

references to iterations and iterative processes

No Comments

No Known Activations