INDEX

Explanations

references to the concept of "unlimited" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

zan

-0.16

affer

-0.15

inel

-0.15

askan

-0.15

ikler

-0.15

ALES

-0.14

å¿ĺ

-0.14

_QUAL

-0.14

ÑĢÑĥÐ½

-0.14

optera

-0.14

POSITIVE LOGITS

/un

0.16

nes

0.16

iesz

0.16

anks

0.14

ness

0.14

ming

0.14

CCR

0.14

çĦ¶

0.14

ating

0.14

unda

0.14

Activations Density 0.004%

references to the concept of "unlimited" in various contexts

No Comments

No Known Activations