INDEX

Explanations

technical terms related to data centers

oai_token-act-pair · gpt-3.5-turbo

references to data centers

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

âķĲ

-0.76

 tremend

-0.75

 piercing

-0.72

shire

-0.71

 destro

-0.65

 erect

-0.63

 unchecked

-0.62

ducers

-0.61

æĢ

-0.61

 wrought

-0.60

POSITIVE LOGITS

abases

1.39

etime

0.98

amation

0.92

Dat

0.91

eman

0.86

dat

0.85

aclysm

0.85

ulia

0.84

emen

0.84

ihad

0.84

Activations Density 0.019%

technical terms related to data centers

references to data centers

No Comments

No Known Activations