INDEX

Explanations

words related to deep exploration or investigation

oai_token-act-pair · gpt-3.5-turbo

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

Ĥİ

-0.77

 royalty

-0.67

 corpus

-0.66

WS

-0.63

Haf

-0.60

 Adobe

-0.60

 ACTIONS

-0.60

 Karin

-0.60

wcsstore

-0.59

Lot

-0.59

POSITIVE LOGITS

hig

1.02

xit

0.94

rer

0.86

rers

0.82

iasm

0.79

ulhu

0.79

ãĥ¥

0.78

rossover

0.76

ghan

0.76

ciating

0.75

Activations Density 0.024%

words related to deep exploration or investigation

No Comments

No Known Activations