INDEX

Explanations

phrases related to written columns or opinion pieces in publications

oai_token-act-pair · gpt-3.5-turbo

references to columns and opinion pieces in publications

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

 embryos

-0.72

 intermedi

-0.71

 embryo

-0.68

 controls

-0.67

 levers

-0.65

wagen

-0.63

lace

-0.61

 maize

-0.61

chrom

-0.60

artifacts

-0.60

POSITIVE LOGITS

 penned

0.94

blogs

0.92

 essays

0.87

 essay

0.85

 columnist

0.81

 commentary

0.80

 titled

0.79

 satir

0.77

 editorial

0.75

 headlined

0.75

Activations Density 0.135%

phrases related to written columns or opinion pieces in publications

references to columns and opinion pieces in publications

No Comments

No Known Activations