INDEX

Explanations

mentions of time durations or specific time periods

oai_token-act-pair · gpt-3.5-turbo

references to time durations and measurements

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

ENC

-0.68

successfully

-0.68

fortune

-0.67

enegger

-0.62

risome

-0.61

POL

-0.60

 optics

-0.60

 contrace

-0.59

 sorts

-0.57

 necessity

-0.57

POSITIVE LOGITS

undred

0.85

 increments

0.82

ousand

0.72

elfth

0.69

abyte

0.68

umbo

0.66

 Wonders

0.66

 Korra

0.65

otropic

0.64

iste

0.64

Activations Density 0.099%

mentions of time durations or specific time periods

references to time durations and measurements

No Comments

No Known Activations