INDEX

Explanations

numeric expressions of time, mainly referring to years, months, weeks, and days in the past

oai_token-act-pair · gpt-3.5-turbo

time-related references indicating a passage of time

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

UCT

-0.65

roit

-0.64

 transporter

-0.61

ymes

-0.61

agn

-0.60

YES

-0.60

MT

-0.59

ework

-0.58

duct

-0.58

nels

-0.57

POSITIVE LOGITS

ago

1.12

Ago

0.88

pring

0.87

hift

0.76

 overdue

0.71

 abroad

0.67

shy

0.65

oshenko

0.65

 evenings

0.64

 farther

0.64

Activations Density 0.090%

numeric expressions of time, mainly referring to years, months, weeks, and days in the past

time-related references indicating a passage of time

No Comments

No Known Activations