INDEX

Explanations

references to a lack of clarity in information or communication

oai_token-act-pair · gpt-3.5-turbo

occurrences of the concept of clarity

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Neighbor

-0.72

esp

-0.71

eder

-0.70

DI

-0.65

 illegally

-0.64

hers

-0.64

vest

-0.63

 Neigh

-0.63

 Lung

-0.63

hern

-0.63

POSITIVE LOGITS

 clarity

3.88

 clarification

1.86

 clearer

1.75

 consistency

1.66

 transparency

1.59

 certainty

1.56

 simplicity

1.52

 purity

1.45

 specificity

1.44

 nuance

1.44

Activations Density 0.012%

references to a lack of clarity in information or communication

occurrences of the concept of clarity

No Comments

No Known Activations