INDEX

Explanations

words related to time comparisons, specifically focusing on changes from a previous specific time period

oai_token-act-pair · gpt-3.5-turbo

mentions of a timeframe that focuses on recent events

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

 Shutterstock

-0.77

cil

-0.75

/#

-0.65

 Reborn

-0.64

 Controls

-0.62

rored

-0.61

aunders

-0.61

Osc

-0.60

CRE

-0.59

 Swords

-0.58

POSITIVE LOGITS

ebin

0.95

 week

0.93

 gasp

0.89

 month

0.88

fm

0.86

 year

0.85

 ditch

0.82

decade

0.82

 rites

0.79

 WEEK

0.76

Activations Density 0.041%

words related to time comparisons, specifically focusing on changes from a previous specific time period

mentions of a timeframe that focuses on recent events

No Comments

No Known Activations