INDEX

Explanations

references to specific roles or identities of individuals within a narrative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 prÃ³p

-0.23

own

-0.22

 same

-0.22

 propio

-0.18

 mesma

-0.18

Own

-0.18

 propia

-0.16

èĩªå·±çļĦ

-0.16

Own

-0.16

 stesso

-0.16

POSITIVE LOGITS

es

0.21

idade

0.20

 dess

0.18

ha

0.17

iÃ³n

0.17

as

0.17

ed

0.15

chaft

0.15

ivalence

0.15

os

0.14

Activations Density 0.032%

references to specific roles or identities of individuals within a narrative context

No Comments

No Known Activations