INDEX

Explanations

words related to measurement, explanation, acknowledgment, and endorsement

oai_token-act-pair · gpt-3.5-turbo

terms related to measurement, acknowledgment, and explanation

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

iencies

-0.76

ãĥĥãĥī

-0.72

iery

-0.71

bold

-0.68

uate

-0.68

estones

-0.67

enic

-0.67

shape

-0.67

avis

-0.66

ells

-0.65

POSITIVE LOGITS

eering

1.01

 thereof

0.92

ary

0.91

ally

0.80

of

0.76

naire

0.76

 process

0.76

ItemTracker

0.76

xual

0.73

alist

0.72

Activations Density 0.220%

words related to measurement, explanation, acknowledgment, and endorsement

terms related to measurement, acknowledgment, and explanation

No Comments

No Known Activations