INDEX

Explanations

phrases or sentences that introduce a new idea or aspect to a conversation

oai_token-act-pair · gpt-3.5-turbo

affirmations or expressions of agreement

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

İĭ

-0.70

 flair

-0.64

âĹ¼

-0.64

illary

-0.63

unal

-0.62

MX

-0.61

 Parables

-0.60

 wedge

-0.58

 lightsaber

-0.58

 garage

-0.58

POSITIVE LOGITS

esley

0.99

come

0.89

ington

0.78

ards

0.76

espie

0.76

Coin

0.68

FTWARE

0.68

tenance

0.67

STON

0.67

 suited

0.67

Activations Density 0.024%

phrases or sentences that introduce a new idea or aspect to a conversation

affirmations or expressions of agreement

No Comments

No Known Activations