INDEX

Explanations

phrases related to cause and effect or explanation

oai_token-act-pair · gpt-3.5-turbo

instances of the word "this."

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

 Pets

-0.70

 Drops

-0.67

agi

-0.65

 Papers

-0.64

 Daniels

-0.64

Personal

-0.63

oots

-0.62

mates

-0.62

aws

-0.61

Fit

-0.61

POSITIVE LOGITS

 particular

0.96

 latter

0.94

 trope

0.88

 phenomenon

0.82

 article

0.80

 newfound

0.79

 arrangement

0.78

 subset

0.78

 invention

0.75

 behaviour

0.74

Activations Density 0.270%

phrases related to cause and effect or explanation

instances of the word "this."

No Comments

No Known Activations