INDEX

Explanations

terms related to mapping and specific keywords like "map," "MAP," and "mapping."

oai_token-act-pair · gpt-3.5-turbo

terms related to mapping and data representation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

aneous

-0.77

nuts

-0.77

ndra

-0.71

acies

-0.70

ocker

-0.64

ews

-0.64

ruciating

-0.63

uates

-0.63

egal

-0.63

Rate

-0.62

POSITIVE LOGITS

leton

0.75

map

0.73

pocket

0.73

iq

0.73

nil

0.71

oly

0.70

riz

0.69

heny

0.68

och

0.67

Map

0.67

Activations Density 0.010%

terms related to mapping and specific keywords like "map," "MAP," and "mapping."

terms related to mapping and data representation

No Comments

No Known Activations