INDEX

Explanations

phrases related to validation processes or actions

oai_token-act-pair · gpt-3.5-turbo

terms related to the concept of validation

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

eworld

-0.87

bill

-0.81

emade

-0.75

ulhu

-0.74

nesses

-0.72

bestos

-0.72

oxin

-0.70

ness

-0.70

ened

-0.70

leness

-0.69

POSITIVE LOGITS

 validation

0.85

 Testing

0.71

idated

0.70

ATES

0.68

ators

0.67

aution

0.67

 Judd

0.65

hurst

0.65

 suites

0.64

 validate

0.63

Activations Density 0.028%

phrases related to validation processes or actions

terms related to the concept of validation

No Comments

No Known Activations