INDEX

Explanations

terms related to time or time periods

oai_token-act-pair · gpt-3.5-turbo

mentions of duration or time-related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

riad

-0.85

inis

-0.77

illard

-0.73

ramid

-0.72

tical

-0.68

leck

-0.67

artisan

-0.67

rim

-0.67

iasis

-0.66

bold

-0.66

POSITIVE LOGITS

 duration

1.31

duration

1.01

Duration

1.00

etheless

0.90

 Duration

0.90

 veter

0.88

 confir

0.84

 destro

0.82

 Ukrain

0.81

dur

0.81

Activations Density 0.004%

terms related to time or time periods

mentions of duration or time-related concepts

No Comments

No Known Activations