INDEX

Explanations

references to preceding content or prior mentions in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

 Discipline

-0.15

 stripslashes

-0.15

wich

-0.15

luck

-0.15

sat

-0.15

mgr

-0.15

ama

-0.14

uers

-0.14

amiliar

-0.14

_Abstract

-0.13

POSITIVE LOGITS

/down

0.15

Ð³Ð¾Ð´

0.15

onec

0.15

 Chore

0.14

anton

0.13

hone

0.13

Invariant

0.13

alÄ±k

0.13

/about

0.13

.documentation

0.13

Activations Density 0.017%

references to preceding content or prior mentions in the text

No Comments

No Known Activations

references to preceding content or prior mentions in the text

No Comments

No Known Activations