INDEX

Explanations

keywords related to speed, growth, or progress

oai_token-act-pair · gpt-3.5-turbo

terms associated with acceleration or speed

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Straw

-0.84

Sut

-0.82

Sew

-0.79

 Flor

-0.74

ief

-0.72

Nab

-0.71

ghan

-0.69

iety

-0.68

 lore

-0.67

sew

-0.67

POSITIVE LOGITS

 accelerated

3.31

 acceleration

3.17

 accelerate

3.07

 accelerating

2.81

 acceler

2.63

 Acceler

2.58

 accelerator

2.55

celer

1.99

 speeding

1.32

 sped

1.31

Activations Density 0.025%

keywords related to speed, growth, or progress

terms associated with acceleration or speed

No Comments

No Known Activations