INDEX

Explanations

instances of objects or people physically seen or found by someone

oai_token-act-pair · gpt-3.5-turbo

instances of the word "spotted" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

nesia

-0.89

pires

-0.83

vernment

-0.81

industrial

-0.79

aceutical

-0.78

mort

-0.77

taboola

-0.77

externalActionCode

-0.77

bern

-0.73

ayers

-0.72

POSITIVE LOGITS

bugs

0.87

owl

0.85

 sightings

0.79

 spotted

0.78

lights

0.75

 Viper

0.72

Owl

0.72

ãĤĬ

0.71

flies

0.69

-+-+

0.69

Activations Density 0.030%

instances of objects or people physically seen or found by someone

instances of the word "spotted" in various contexts

No Comments

No Known Activations