INDEX

Explanations

instances of the word "when" and phrases that indicate conditional scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

ye

-0.19

mach

-0.16

ç¯Ģ

-0.15

illez

-0.15

abit

-0.15

 stren

-0.14

ãģ«ãģĭ

-0.14

anie

-0.14

ober

-0.14

makt

-0.13

POSITIVE LOGITS

CLUD

0.16

GMEM

0.16

ollision

0.15

/|

0.15

ampo

0.15

ometown

0.15

ÏĢÎ·

0.14

 baÅŁÄ±na

0.14

celik

0.14

 drag

0.14

Activations Density 0.247%

instances of the word "when" and phrases that indicate conditional scenarios

No Comments

No Known Activations