INDEX

Explanations

time periods specified in months or weeks

oai_token-act-pair · gpt-3.5-turbo

phrases indicating specific time periods or durations

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

 unfocusedRange

-0.75

 srfAttach

-0.73

Rated

-0.73

ulkan

-0.72

interrupted

-0.70

conservancy

-0.69

ictive

-0.67

bringing

-0.63

owship

-0.62

recomm

-0.62

POSITIVE LOGITS

imester

1.11

eve

0.73

 Territ

0.70

 Coinbase

0.68

 arrivals

0.68

McA

0.68

 Archie

0.67

 Grayson

0.67

 Huck

0.67

 whiff

0.67

Activations Density 0.083%

time periods specified in months or weeks

phrases indicating specific time periods or durations

No Comments

No Known Activations