INDEX

Explanations

phrases that indicate topics or subjects of discussion

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

udge

-0.15

 UserRole

-0.14

 demand

-0.14

udic

-0.14

\brief

-0.14

vero

-0.14

 Riders

-0.14

ÅĻiv

-0.14

ÎºÎŃ

-0.14

kova

-0.14

POSITIVE LOGITS

0.15

omi

0.14

imd

0.14

ngh

0.13

els

0.13

PN

0.13

staking

0.13

tsy

0.13

å¸ģ

0.13

Activations Density 0.079%

phrases that indicate topics or subjects of discussion

No Comments

No Known Activations