INDEX

Explanations

quantifiable claims or questions that imply evidence or statistics

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

ÅĻÃŃd

-0.16

_Lean

-0.15

ÌĨ

-0.14

avou

-0.14

/cmd

-0.13

antas

-0.13

oÅ¾

-0.13

utomation

-0.13

ÐµÐ»Ð¸Ðº

-0.13

å¦

-0.13

POSITIVE LOGITS

ë²

0.14

Ã¶yle

0.14

0.13

umann

0.13

_rng

0.13

uns

0.13

par

0.13

spiel

0.13

MOVED

0.13

dit

0.13

Activations Density 0.044%

quantifiable claims or questions that imply evidence or statistics

No Comments

No Known Activations