INDEX

Explanations

occurrences of the word "Happy"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Happy"

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

arin

-0.81

resent

-0.74

rovers

-0.73

aeda

-0.73

$$$$

-0.73

IDER

-0.72

enses

-0.72

urally

-0.71

ĳ

-0.70

ental

-0.69

POSITIVE LOGITS

 Birthday

0.88

 Happy

0.86

Happy

0.84

 birthday

0.82

joy

0.80

ness

0.80

 Gilmore

0.77

balls

0.74

 Meal

0.73

bies

0.72

Activations Density 0.017%

occurrences of the word "Happy"

occurrences of the word "Happy"

No Comments

No Known Activations