INDEX

Explanations

phrases implying contrasting viewpoints or actions

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Juf

-0.97

aen

-0.94

NOO

-0.93

 thut

-0.90

 „,

-0.90

 Hano

-0.90

ftu

-0.88

 Febru

-0.88

ufe

-0.88

 nomine

-0.88

POSITIVE LOGITS

<bos>

0.58

AccessorTable

0.53

 desire

0.52

bidden

0.49

ември

0.49

USTAIN

0.48

 vlieg

0.48

ValueStyle

0.48

AppRoutingModule

0.47

Necesito

0.46

Activations Density 0.146%

phrases implying contrasting viewpoints or actions

No Comments

No Known Activations