INDEX

Explanations

names of individuals and family relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

compact

-0.15

folio

-0.15

 compact

-0.15

à¸±à¸Ħà¸£

-0.14

plorer

-0.14

fon

-0.14

apl

-0.14

ëĭ¤ê°Ģ

-0.14

ncoder

-0.14

lej

-0.14

POSITIVE LOGITS

 whom

0.22

 preceded

0.19

preced

0.19

 deceased

0.16

ofs

0.16

ãĢĤãĢĤ↵↵

0.14

oping

0.14

ceased

0.14

 survives

0.14

 survive

0.14

Activations Density 0.017%

names of individuals and family relationships

No Comments

No Known Activations