INDEX

Explanations

measurements or quantities

oai_token-act-pair · gpt-3.5-turbo

terms related to measurement and statistical evaluation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

lez

-0.75

sle

-0.72

herty

-0.71

olla

-0.69

 Reserve

-0.68

 Safari

-0.68

TOR

-0.67

burning

-0.67

ipeg

-0.67

yles

-0.65

POSITIVE LOGITS

asure

0.77

ment

0.76

urgy

0.75

eters

0.75

urgical

0.74

 measuring

0.74

agen

0.72

MENTS

0.72

posts

0.71

asured

0.70

Activations Density 0.049%

measurements or quantities

terms related to measurement and statistical evaluation

No Comments

No Known Activations