INDEX

Explanations

time-related expressions, specifically durations or lengths of time

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the duration of time or length of activities

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

 inher

-0.62

 inherit

-0.59

ELL

-0.58

ãĤ¨ãĥ«

-0.57

ãĥķãĤ©

-0.56

 Eleanor

-0.56

 inherited

-0.55

 reformed

-0.55

 induct

-0.55

 Redmond

-0.54

POSITIVE LOGITS

 minutes

1.04

 hour

1.02

 hours

0.98

minute

0.96

 seconds

0.96

hours

0.95

seconds

0.92

 minute

0.90

hour

0.86

hrs

0.85

Activations Density 0.299%

time-related expressions, specifically durations or lengths of time

phrases indicating the duration of time or length of activities

No Comments

No Known Activations