INDEX

Explanations

prepositions and their usage within context

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

ndx

-0.15

Americ

-0.15

anger

-0.14

 belang

-0.14

Ð¾ÑĢÐ¾Ð¶

-0.14

Ð¼ÐµÐ½ÑĪ

-0.14

arefa

-0.14

itize

-0.14

attery

-0.14

nat

-0.14

POSITIVE LOGITS

onu

0.15

sett

0.15

jax

0.14

ile

0.13

-transitional

0.13

odos

0.13

Ð¸Ð¼Ð°

0.13

 help

0.13

iled

0.13

keh

0.13

Activations Density 0.280%

prepositions and their usage within context

No Comments

No Known Activations