INDEX

Explanations

personal pronouns and expressions of self-reference

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

strom

-0.16

rack

-0.15

maker

-0.15

-SA

-0.15

kud

-0.14

nite

-0.14

 Vance

-0.14

stime

-0.14

tak

-0.14

Ð¾Ð¼ÐµÑĢ

-0.14

POSITIVE LOGITS

0.16

å¡ļ

0.14

 stab

0.14

loquent

0.14

thers

0.14

Ged

0.14

_hello

0.14

OKEN

0.14

 setHidden

0.14

VEC

0.13

Activations Density 0.025%

personal pronouns and expressions of self-reference

No Comments

No Known Activations