INDEX

Explanations

terms related to the concept of "one" or "single" elements among 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 @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

Ã¼z

-0.16

ekim

-0.14

ichert

-0.14

aina

-0.14

Hen

-0.14

rant

-0.14

TAIL

-0.14

.hm

-0.14

åĵģ

-0.14

 ourselves

-0.14

POSITIVE LOGITS

ÃĹ↵↵

0.15

_PLUS

0.14

 Ð·Ð°ÑģÑĤ

0.14

//{{

0.14

atest

0.14

---</

0.14

ertiary

0.14

elson

0.13

WithURL

0.13

.dtd

0.13

Activations Density 0.037%

terms related to the concept of "one" or "single" elements among various contexts

No Comments

No Known Activations