INDEX

Explanations

numerical references or identifiers related to articles or studies

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

errick

-0.15

 Morrow

-0.15

 Meer

-0.15

ault

-0.15

_retry

-0.15

quet

-0.15

ohen

-0.14

 ç·

-0.14

ÅĤad

-0.14

Disposition

-0.14

POSITIVE LOGITS

 otherwise

0.17

ager

0.17

ro

0.15

 Carlton

0.14

 atol

0.14

lap

0.14

_Begin

0.14

uttle

0.14

 stan

0.13

Activations Density 0.000%

numerical references or identifiers related to articles or studies

No Comments

No Known Activations