INDEX

Explanations

instances of strong emotional expressions or affirmations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

zan

-0.16

yer

-0.16

rang

-0.15

anka

-0.14

ittest

-0.14

dig

-0.14

elman

-0.14

bid

-0.14

utos

-0.13

dem

-0.13

POSITIVE LOGITS

ussy

0.16

ocket

0.16

0.14

yon

0.14

vn

0.14

obus

0.14

áº£n

0.13

spb

0.13

lyn

0.13

andle

0.13

Activations Density 0.000%

instances of strong emotional expressions or affirmations

No Comments

No Known Activations