INDEX

Explanations

dates or time-related words preceded by "last"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "last" followed by time periods such as weeks, months, and years

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

holes

-0.78

cil

-0.74

aunders

-0.71

olution

-0.67

Osc

-0.66

olutions

-0.64

death

-0.63

/#

-0.62

ysc

-0.60

oidal

-0.59

POSITIVE LOGITS

 month

1.12

 week

1.09

 year

1.02

 night

0.93

 summer

0.90

 ditch

0.87

 semester

0.86

yr

0.84

 Month

0.84

 winter

0.83

Activations Density 0.065%

dates or time-related words preceded by "last"

instances of the word "last" followed by time periods such as weeks, months, and years

No Comments

No Known Activations