INDEX

Explanations

instances of the word "scaling" and related terms

oai_token-act-pair · gpt-3.5-turbo

terms related to scaling or increasing capacity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

vous

-0.71

odor

-0.68

uala

-0.66

ech

-0.66

algia

-0.65

esson

-0.65

odge

-0.65

icipated

-0.64

riad

-0.64

essler

-0.64

POSITIVE LOGITS

Factor

0.84

itized

0.81

 heights

0.81

 scale

0.79

 invari

0.75

 scales

0.74

 scaled

0.72

 ratios

0.72

Scale

0.70

 adjustment

0.69

Activations Density 0.026%

instances of the word "scaling" and related terms

terms related to scaling or increasing capacity

No Comments

No Known Activations