INDEX

Explanations

descriptions of different approaches or methods

oai_token-act-pair · gpt-3.5-turbo

mentions of different approaches or methodologies

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

watching

-0.74

cakes

-0.74

gin

-0.71

arus

-0.70

cake

-0.70

rake

-0.69

ãĥ©ãĥ³

-0.68

Wak

-0.68

ongo

-0.68

ensen

-0.67

POSITIVE LOGITS

 approach

0.94

 Approach

0.88

ahime

0.78

idon

0.74

 approaches

0.71

rait

0.71

lectic

0.70

olitan

0.70

perty

0.70

oteric

0.70

Activations Density 0.029%

descriptions of different approaches or methods

mentions of different approaches or methodologies

No Comments

No Known Activations