INDEX

Explanations

phrases related to concepts of truth, lies, and power dynamics

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

ackers

-0.15

rame

-0.15

ervo

-0.15

coli

-0.15

ãģ¨ãģĨ

-0.15

/sdk

-0.15

xp

-0.14

elu

-0.14

elong

-0.14

atee

-0.14

POSITIVE LOGITS

 Bent

0.15

ewn

0.15

 Correction

0.15

ÅĪ

0.14

 Physician

0.14

 transport

0.14

Ã¥r

0.14

as

0.14

pe

0.14

Activations Density 0.001%

phrases related to concepts of truth, lies, and power dynamics

No Comments

No Known Activations