INDEX

Explanations

instances where actions or events cause consequences or reactions

oai_token-act-pair · gpt-3.5-turbo

references to causes and their consequences

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

lished

-0.96

ACTED

-0.72

ORTS

-0.70

 sugg

-0.70

ablish

-0.70

ADRA

-0.70

 Provided

-0.68

OLOGY

-0.67

ocally

-0.67

razen

-0.66

POSITIVE LOGITS

 havoc

1.25

 headaches

0.89

 disturbance

0.86

 damage

0.86

 mayhem

0.84

 wedge

0.80

 harm

0.79

 wrath

0.79

 unnecessary

0.78

 disruption

0.75

Activations Density 0.137%

instances where actions or events cause consequences or reactions

references to causes and their consequences

No Comments

No Known Activations