INDEX

Explanations

punctuation and phrases that indicate conversational or emotional pauses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

asso

-0.17

qus

-0.16

ovah

-0.15

anki

-0.15

alet

-0.15

rych

-0.15

acob

-0.14

.UIManager

-0.14

-www

-0.14

antha

-0.14

POSITIVE LOGITS

0.15

 Subscriber

0.14

uka

0.14

DataContract

0.14

 Maher

0.14

_esc

0.14

ahn

0.13

chaft

0.13

Subscriber

0.13

Activations Density 0.001%

punctuation and phrases that indicate conversational or emotional pauses

No Comments

No Known Activations