INDEX

Explanations

mentions of machine learning and related terms

oai_token-act-pair · gpt-3.5-turbo

terms and concepts related to machine learning

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

 Daylight

-0.75

 Okin

-0.73

swick

-0.72

irgin

-0.70

 Sinclair

-0.66

 Kuwait

-0.64

nep

-0.64

ufact

-0.63

tar

-0.63

bestos

-0.61

POSITIVE LOGITS

 algorithms

1.34

 algorithm

1.10

 computation

0.86

 neural

0.85

 optimization

0.84

 architectures

0.84

 algorith

0.82

 computational

0.81

 bots

0.78

 processor

0.78

Activations Density 0.072%

mentions of machine learning and related terms

terms and concepts related to machine learning

No Comments

No Known Activations