INDEX

Explanations

specific occurrences of unique characters and function names in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

uala

-0.16

igkeit

-0.15

ignon

-0.15

-ajax

-0.15

_LS

-0.14

iger

-0.14

_globals

-0.14

 Neville

-0.14

 SetValue

-0.14

ima

-0.14

POSITIVE LOGITS

Sm

0.21

 Smoke

0.20

 Smith

0.19

/sm

0.19

Sm

0.19

smith

0.19

SM

0.18

sm

0.18

 smokers

0.17

(sm

0.17

Activations Density 0.005%

specific occurrences of unique characters and function names in the text

No Comments

No Known Activations