INDEX

Explanations

terms related to landscapes

oai_token-act-pair · gpt-3.5-turbo

references to landscapes, particularly in contexts related to environment, geography, or settings

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

unker

-0.93

ibo

-0.85

uana

-0.83

etts

-0.83

uster

-0.80

ulate

-0.78

cial

-0.75

idel

-0.74

thood

-0.74

ule

-0.74

POSITIVE LOGITS

scape

1.04

 landscapes

0.88

 Rover

0.86

 vegetation

0.81

 landscape

0.80

 backdrop

0.75

ttes

0.74

 geometry

0.72

 scenery

0.71

 gardens

0.70

Activations Density 0.025%

terms related to landscapes

references to landscapes, particularly in contexts related to environment, geography, or settings

No Comments

No Known Activations