INDEX

Explanations

words related to verification or confirmation

oai_token-act-pair · gpt-3.5-turbo

terms related to validation processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Anth

-0.86

Tus

-0.74

ellig

-0.71

enta

-0.71

Sus

-0.71

 Keeper

-0.71

 slain

-0.70

Effective

-0.70

atha

-0.68

aryl

-0.67

POSITIVE LOGITS

 validation

2.25

 validate

1.55

opped

1.55

 validated

1.49

valid

1.44

 pity

1.36

 trivial

1.30

 nostalgia

1.29

 McGregor

1.29

opping

1.21

Activations Density 0.040%

words related to verification or confirmation

terms related to validation processes

No Comments

No Known Activations