INDEX

Explanations

terms related to evaluating or analyzing expressions in programming

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

akest

-0.17

rawer

-0.16

arella

-0.16

PointF

-0.15

iosper

-0.15

Ã³ng

-0.14

esian

-0.14

_skb

-0.14

Ú©ÙĪ

-0.14

ÃĹ↵↵

-0.14

POSITIVE LOGITS

nia

0.14

ENCHMARK

0.14

OND

0.14

iken

0.14

pte

0.14

é¼»

0.14

ila

0.14

 stÃ¡tnÃŃ

0.14

otch

0.13

Activations Density 0.008%

terms related to evaluating or analyzing expressions in programming

No Comments

No Known Activations