INDEX

Explanations

conjunctions and adverbs indicating conditionality or interconnected ideas

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

immer

-0.16

ÑģÑĤÐµÑĢ

-0.15

ceased

-0.15

posable

-0.15

ovÃ¡ny

-0.15

AILS

-0.14

LEAN

-0.14

Nam

-0.14

 distance

-0.14

DBG

-0.14

POSITIVE LOGITS

icon

0.17

ral

0.16

ever

0.15

ning

0.15

Wal

0.14

lu

0.14

 icon

0.14

 Ping

0.14

'icon

0.14

rob

0.13

Activations Density 0.363%

conjunctions and adverbs indicating conditionality or interconnected ideas

No Comments

No Known Activations