INDEX

Explanations

proper nouns related to intelligence agencies and investigations

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

 كومونز

-0.70

 unlaw

-0.59

 superintend

-0.59

 liberality

-0.58

 gratify

-0.58

 friable

-0.57

 tolerably

-0.57

 mortgagee

-0.56

ougars

-0.56

 unspeak

-0.54

POSITIVE LOGITS

Cla

1.66

Cla

1.58

cla

1.47

cla

1.36

CLA

1.22

CLA

1.01

 Clare

0.93

kla

0.87

Clare

0.87

 claws

0.85

Activations Density 0.115%

proper nouns related to intelligence agencies and investigations

No Comments

No Known Activations