INDEX

Explanations

references to or descriptions of dwarfs/dwarves

oai_token-act-pair · gpt-3.5-turbo

references to dwarfs or dwarves 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 @ 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

orie

-0.85

inventoryQuantity

-0.79

ettings

-0.75

ored

-0.71

 Clip

-0.71

mble

-0.68

essee

-0.68

 Seller

-0.66

 antioxid

-0.66

henko

-0.65

POSITIVE LOGITS

 Dwar

0.89

 Dwarf

0.87

 Fortress

0.84

 dwarves

0.83

 dwarf

0.82

ishly

0.81

 elves

0.80

shire

0.78

sson

0.72

haus

0.70

Activations Density 0.017%

references to or descriptions of dwarfs/dwarves

references to dwarfs or dwarves in various contexts

No Comments

No Known Activations