INDEX

Explanations

the word "nasty" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

led

-0.16

ãģ¾ãģ¾

-0.15

ilda

-0.15

ultipart

-0.15

ategory

-0.14

ãĤ¹ãĥĿ

-0.14

NAS

-0.14

pling

-0.14

uggage

-0.13

.flip

-0.13

POSITIVE LOGITS

RowCount

0.16

jax

0.15

 ëħĦ

0.14

arf

0.14

mons

0.14

 Accountability

0.14

_dup

0.14

se

0.13

anz

0.13

Activations Density 0.006%

the word "nasty" and its variations

No Comments

No Known Activations