INDEX

Explanations

specific numerical references, likely related to research citations or DOI links

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

rif

-0.15

ikat

-0.14

antt

-0.14

nap

-0.14

estre

-0.13

orate

-0.13

piar

-0.13

irates

-0.13

Äįit

-0.13

prech

-0.13

POSITIVE LOGITS

addock

0.16

ìĶ

0.15

://

0.15

aths

0.15

abelle

0.14

_OPTS

0.14

ä¹ħ

0.14

ABEL

0.14

oeff

0.14

ehler

0.14

Activations Density 0.012%

specific numerical references, likely related to research citations or DOI links

No Comments

No Known Activations