INDEX

Explanations

legal and organizational terms and concepts

oai_token-act-pair · gpt-3.5-turbo

terms related to formal documents and processes

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

Redd

-0.74

Diamond

-0.69

Brit

-0.63

Ty

-0.63

Blood

-0.62

Tik

-0.61

Char

-0.59

ãĥ©ãĥ³

-0.58

 Blend

-0.58

onz

-0.57

POSITIVE LOGITS

 itself

1.13

eers

0.76

 amounted

0.74

 consisted

0.72

 culmin

0.71

 designation

0.71

's

0.68

naires

0.68

 himself

0.68

 portion

0.68

Activations Density 0.426%

legal and organizational terms and concepts

terms related to formal documents and processes

No Comments

No Known Activations