INDEX

Explanations

numeric rankings and positions in competitive contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

olle

-0.16

ritz

-0.16

aeda

-0.15

-0.14

ulur

-0.14

klu

-0.13

remote

-0.13

%X

-0.13

çŃĭ

-0.13

POSITIVE LOGITS

afari

0.16

ople

0.16

å¸

0.14

pir

0.14

oon

0.14

ictor

0.14

ìĭ

0.14

_mD

0.14

yme

0.13

TAIL

0.13

Activations Density 0.046%

numeric rankings and positions in competitive contexts

No Comments

No Known Activations

numeric rankings and positions in competitive contexts

No Comments

No Known Activations