INDEX

Explanations

references to columns in data structures

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

lemen

-0.18

ëģĶ

-0.16

oward

-0.15

riel

-0.15

-ÑĤÐ¾

-0.15

lings

-0.15

porto

-0.15

ÑĢÐ¾Ð²Ð¸Ñĩ

-0.14

downloads

-0.14

eacher

-0.14

POSITIVE LOGITS

arity

0.20

aires

0.18

ãĤº

0.16

ÑģÑĮ

0.15

idable

0.15

nect

0.15

aments

0.15

aliz

0.15

naires

0.15

peÄį

0.15

Activations Density 0.057%

references to columns in data structures

No Comments

No Known Activations