INDEX

Explanations

words related to workshops and events

oai_token-act-pair · gpt-3.5-turbo

references to workshops

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

judicial

-0.67

ties

-0.65

sec

-0.65

croft

-0.64

yll

-0.64

bis

-0.62

rises

-0.62

iod

-0.62

pse

-0.61

thus

-0.60

POSITIVE LOGITS

 workshop

1.14

smanship

1.07

 Workshop

1.04

 workshops

0.96

dayName

0.87

hops

0.85

workshop

0.84

craft

0.81

ItemTracker

0.80

halla

0.79

Activations Density 0.012%

words related to workshops and events

references to workshops

No Comments

No Known Activations