INDEX

Explanations

specific phrases related to cause and effect or results

oai_token-act-pair · gpt-3.5-turbo

questions and statements that present results or conclusions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

igraph

-0.79

yi

-0.71

 mant

-0.70

ancies

-0.63

inventoryQuantity

-0.63

addons

-0.63

igion

-0.59

 mantle

-0.58

 branch

-0.58

 Antiqu

-0.58

POSITIVE LOGITS

 predictably

0.83

ãĤª

0.75

 remarkably

0.74

 predictable

0.74

 disastrous

0.74

 hilarious

0.72

 astounding

0.71

 priceless

0.69

 astonishing

0.67

 spectacular

0.66

Activations Density 0.156%

specific phrases related to cause and effect or results

questions and statements that present results or conclusions

No Comments

No Known Activations