INDEX

Explanations

locations or settings within maps

oai_token-act-pair · gpt-3.5-turbo

references to gameplay maps

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

icer

-0.94

ilipp

-0.88

æł

-0.80

nuts

-0.77

nder

-0.77

ndra

-0.76

tymology

-0.76

elvet

-0.75

orld

-0.73

idth

-0.72

POSITIVE LOGITS

ographer

0.78

 tiles

0.77

ãĤº

0.76

map

0.72

map

0.72

 Pool

0.71

 sprite

0.70

 maps

0.69

 pool

0.68

front

0.68

Activations Density 0.014%

locations or settings within maps

references to gameplay maps

No Comments

No Known Activations