INDEX

Explanations

terms related to layers or levels of something

oai_token-act-pair · gpt-3.5-turbo

references to layers 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 @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ãĥīãĥ©

-0.84

orie

-0.80

date

-0.78

etheus

-0.77

ILLE

-0.76

 Predators

-0.76

lishing

-0.75

itarian

-0.74

photo

-0.70

speaking

-0.70

POSITIVE LOGITS

 layers

1.30

 layer

1.10

 thickness

1.10

layer

0.80

 veil

0.79

 separ

0.77

 opacity

0.76

 layered

0.75

 coats

0.75

arity

0.74

Activations Density 0.043%

terms related to layers or levels of something

references to layers in various contexts

No Comments

No Known Activations