INDEX

Explanations

mathematical expressions and formulas

oai_token-act-pair · gpt-3.5-turbo

mathematical functions and statistical terms

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

ettings

-0.84

 chats

-0.74

::::::::

-0.71

EVA

-0.70

 menus

-0.70

 nods

-0.69

aug

-0.66

 lobb

-0.66

Attach

-0.66

 safeguards

-0.66

POSITIVE LOGITS

 theorem

1.09

 coefficient

1.05

orem

1.02

frac

1.02

 âĪ

1.01

 ÎĶ

1.00

 inverse

0.97

 ÏĦ

0.96

 computed

0.96

 coefficients

0.94

Activations Density 0.434%

mathematical expressions and formulas

mathematical functions and statistical terms

No Comments

No Known Activations