INDEX

Explanations

references to specific entities or sources, such as individuals, organizations, or publications

oai_token-act-pair · gpt-3.5-turbo

references to visual media, specifically photos and illustrations

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

 Arena

-0.63

 advisory

-0.62

 repeal

-0.62

 interim

-0.61

 probation

-0.60

 consecutive

-0.59

 costing

-0.59

 haven

-0.59

 totaling

-0.58

 hands

-0.58

POSITIVE LOGITS

Sensor

0.85

Newsletter

0.79

Loading

0.75

-+-+

0.74

SPONSORED

0.73

 dstg

0.72

Phot

0.72

Done

0.72

oter

0.72

notations

0.71

Activations Density 0.166%

references to specific entities or sources, such as individuals, organizations, or publications

references to visual media, specifically photos and illustrations

No Comments

No Known Activations