INDEX

Explanations

instances where tasks or actions have been completed

oai_token-act-pair · gpt-3.5-turbo

phrases related to the completion of tasks or challenges

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

maid

-0.74

gypt

-0.67

lua

-0.65

hot

-0.64

spr

-0.63

brow

-0.63

ung

-0.62

bour

-0.59

hee

-0.56

LI

-0.56

POSITIVE LOGITS

teness

0.97

issance

0.89

ances

0.85

TAIN

0.84

 completion

0.77

srf

0.76

 Completed

0.72

taining

0.72

 feasibility

0.71

 preparations

0.70

Activations Density 0.028%

instances where tasks or actions have been completed

phrases related to the completion of tasks or challenges

No Comments

No Known Activations