INDEX

Explanations

significant nouns, particularly those related to people’s experiences and emotions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ona

-0.17

STE

-0.15

ONA

-0.15

Ste

-0.14

ior

-0.14

Inlining

-0.14

ÐµÑģÐ°

-0.14

_tunnel

-0.14

TECTED

-0.14

POSITIVE LOGITS

asil

0.17

ombat

0.15

ÑĩÐµ

0.15

kowski

0.15

Ã¹ng

0.15

 Doyle

0.15

ehr

0.14

 Pell

0.14

veal

0.14

oud

0.14

Activations Density 0.002%

significant nouns, particularly those related to people’s experiences and emotions

No Comments

No Known Activations