INDEX

Explanations

statements related to presenting or analyzing evidence in various contexts

oai_token-act-pair · gpt-3.5-turbo

phrases related to the presence and support of evidence

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

ahime

-0.81

ModLoader

-0.71

NESS

-0.70

Bomb

-0.70

Vert

-0.69

 kisses

-0.69

inventoryQuantity

-0.68

ettings

-0.68

 Hispan

-0.67

 Maintenance

-0.67

POSITIVE LOGITS

 corrobor

1.02

evidence

0.99

 eyewitness

0.96

 dispro

0.94

 evidence

0.90

 factual

0.84

 exoner

0.83

AUD

0.83

ewitness

0.81

 substant

0.79

Activations Density 0.214%

statements related to presenting or analyzing evidence in various contexts

phrases related to the presence and support of evidence

No Comments

No Known Activations