INDEX

Explanations

references to themes of danger or threats associated with creatures or characters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

jem

-0.15

cela

-0.15

.restaurant

-0.15

umo

-0.15

azzi

-0.15

 newest

-0.15

èħ

-0.14

STAT

-0.14

ãĥ¼ãĥī

-0.14

verity

-0.14

POSITIVE LOGITS

icious

0.15

aroo

0.15

 survival

0.14

é®

0.14

.rank

0.14

 Rank

0.14

 å¯

0.13

 Survival

0.13

directory

0.13

ERNEL

0.13

Activations Density 0.056%

references to themes of danger or threats associated with creatures or characters

No Comments

No Known Activations