INDEX

Explanations

words or phrases indicating various states, conditions, or attributes, especially those with clear distinctions or identities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

Opaque

-0.15

ibling

-0.15

plural

-0.15

Ø§Ø¨ÙĬ

-0.15

lint

-0.15

ryn

-0.15

env

-0.15

eyn

-0.14

Gib

-0.14

 Grade

-0.14

POSITIVE LOGITS

elts

0.17

iglia

0.16

ç©

0.16

 Mull

0.16

_attached

0.15

arin

0.15

è

0.15

è£ķ

0.15

Kel

0.15

Moh

0.15

Activations Density 0.021%

words or phrases indicating various states, conditions, or attributes, especially those with clear distinctions or identities

No Comments

No Known Activations