INDEX

Explanations

suffixes indicating various philosophical or cultural concepts

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

 Stafford

-0.15

dim

-0.15

cassert

-0.14

arden

-0.14

boxed

-0.14

ì§ĵ

-0.14

æħ§

-0.14

Ã²ng

-0.14

ence

-0.14

lington

-0.13

POSITIVE LOGITS

eren

0.16

.bz

0.15

Gaw

0.15

ptune

0.14

thalm

0.14

.Handled

0.14

vell

0.14

abee

0.14

stadt

0.14

/Dk

0.14

Activations Density 0.165%

suffixes indicating various philosophical or cultural concepts

No Comments

No Known Activations