INDEX

Explanations

technical terms related to data processing and computing, specifically related to loading data

oai_token-act-pair · gpt-3.5-turbo

references to loading or the concept of load in various contexts

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

 Preservation

-0.72

 Noir

-0.70

cour

-0.70

 Paran

-0.69

éĥ

-0.69

 ruled

-0.69

pires

-0.68

Ó

-0.68

venth

-0.65

inton

-0.63

POSITIVE LOGITS

 load

3.95

 loads

2.71

 Load

2.51

load

2.49

 loading

2.16

Load

2.04

loads

1.91

 loaded

1.79

 loader

1.68

LOAD

1.60

Activations Density 0.014%

technical terms related to data processing and computing, specifically related to loading data

references to loading or the concept of load in various contexts

No Comments

No Known Activations