INDEX

Explanations

phrases and words associated with auditory engagement and listening

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

illard

-0.16

ague

-0.16

uls

-0.15

ãĥ³ãĤ¬

-0.15

isted

-0.15

ej

-0.15

reso

-0.14

aget

-0.14

sten

-0.14

POSITIVE LOGITS

/watch

0.19

AndServe

0.17

/view

0.17

 ØŃØ°

0.15

/vnd

0.15

æĻ´

0.15

-only

0.14

 ears

0.14

ascus

0.14

á»ĩn

0.14

Activations Density 0.033%

phrases and words associated with auditory engagement and listening

No Comments

No Known Activations