INDEX

Explanations

references to specific time periods or lengths of time

oai_token-act-pair · gpt-3.5-turbo

references to the concept of duration in various 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

riad

-0.90

ramid

-0.89

otle

-0.86

illard

-0.82

dinand

-0.80

unin

-0.77

iasis

-0.74

tical

-0.74

inatory

-0.71

artisan

-0.71

POSITIVE LOGITS

 duration

1.01

Duration

0.98

duration

0.92

 periods

0.91

 elapsed

0.86

dur

0.83

 lapse

0.82

 ÏĦ

0.80

 intervals

0.78

frames

0.77

Activations Density 0.029%

references to specific time periods or lengths of time

references to the concept of duration in various contexts

No Comments

No Known Activations