INDEX

Explanations

statements made by individuals following specific events or occurrences

oai_token-act-pair · gpt-3.5-turbo

phrases that refer to actions or statements made following an event

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

nexus

-0.75

geist

-0.70

onne

-0.68

itect

-0.65

tw

-0.65

azon

-0.65

cin

-0.65

iman

-0.64

cause

-0.64

omorphic

-0.63

POSITIVE LOGITS

 announcing

1.07

 unveiling

1.03

 briefing

0.92

 signing

0.92

 inspecting

0.90

 congrat

0.85

terday

0.84

 meeting

0.83

 dismissing

0.80

 departing

0.80

Activations Density 0.182%

statements made by individuals following specific events or occurrences

phrases that refer to actions or statements made following an event

No Comments

No Known Activations