INDEX

Explanations

time-related phrases, specifically referencing durations like "months" or "years."

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the passage of time, specifically referencing occurrences in the recent past

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 thence

-0.68

adena

-0.66

icut

-0.66

obal

-0.64

UAL

-0.61

ater

-0.61

obo

-0.60

wrapper

-0.60

aughter

-0.59

 ideally

-0.58

POSITIVE LOGITS

 months

1.41

 weeks

1.40

 years

1.28

 days

1.28

 decades

1.20

 hundred

1.15

months

1.07

 seasons

1.06

 dozen

1.05

 paragraphs

1.05

Activations Density 0.055%

time-related phrases, specifically referencing durations like "months" or "years."

phrases indicating the passage of time, specifically referencing occurrences in the recent past

No Comments

No Known Activations