INDEX

Explanations

phrases related to time durations

oai_token-act-pair · gpt-3.5-turbo

references to time periods or durations expressed in years and distinctions in legal contexts

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

 Rica

-0.75

 Scotia

-0.72

 Borders

-0.69

 Gillespie

-0.69

 Nero

-0.68

 Kafka

-0.68

AVG

-0.67

 Proc

-0.66

 Tort

-0.65

 Richards

-0.65

POSITIVE LOGITS

year

1.36

legged

1.34

month

1.25

sided

1.24

minute

1.22

thirds

1.21

hour

1.20

day

1.20

dimensional

1.20

tailed

1.19

Activations Density 0.079%

phrases related to time durations

references to time periods or durations expressed in years and distinctions in legal contexts

No Comments

No Known Activations