INDEX

Explanations

references to federal government and its regulations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ida

-0.17

cus

-0.16

ce

-0.15

áº¿u

-0.14

bread

-0.14

idas

-0.14

Î¹Î²

-0.14

¬

-0.14

Ã¨res

-0.13

iropr

-0.13

POSITIVE LOGITS

/local

0.18

ized

0.18

/state

0.16

ised

0.15

/world

0.15

ization

0.14

wide

0.14

'gc

0.14

most

0.14

Activations Density 0.021%

references to federal government and its regulations

No Comments

No Known Activations