INDEX

Explanations

references to educational resources, particularly encyclopedic entries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

áº©u

-0.15

utzer

-0.15

Ã¥r

-0.14

ISTA

-0.14

DC

-0.14

Äģn

-0.13

 Bever

-0.13

ement

-0.13

andre

-0.13

POSITIVE LOGITS

onde

0.15

RowAt

0.15

hound

0.14

.dex

0.14

turnstile

0.14

ysi

0.14

 symbolism

0.14

_PROVID

0.13

lopedia

0.13

 Ð¿ÑĸÐ´Ñģ

0.13

Activations Density 0.037%

references to educational resources, particularly encyclopedic entries

No Comments

No Known Activations