INDEX

Explanations

references to specific ingredients or components within a context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

akk

-0.14

ACHE

-0.14

Ne

-0.14

 Neil

-0.14

ichi

-0.14

olec

-0.13

achi

-0.13

 repl

-0.13

unden

-0.13

Neil

-0.13

POSITIVE LOGITS

fall

0.15

iatric

0.14

 particular

0.14

allah

0.14

athed

0.14

eu

0.14

 batt

0.13

adam

0.13

leon

0.13

ìĤ´

0.13

Activations Density 0.055%

references to specific ingredients or components within a context

No Comments

No Known Activations