INDEX

Explanations

instances of the word "little"

oai_token-act-pair · gpt-3.5-turbo

references to "little" entities, particularly in a playful or endearing context

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

anwhile

-0.88

chwitz

-0.85

idential

-0.80

 conclud

-0.80

cius

-0.78

igham

-0.77

ãĥķãĤ©

-0.75

antage

-0.75

arbon

-0.74

worthiness

-0.70

POSITIVE LOGITS

bit

0.99

 girl

0.87

 girls

0.87

boy

0.83

 boys

0.79

kid

0.78

 bits

0.78

 snippets

0.76

 consolation

0.72

 helpers

0.72

Activations Density 0.032%

instances of the word "little"

references to "little" entities, particularly in a playful or endearing context

No Comments

No Known Activations