INDEX

Explanations

terms related to dispatching or sending out responses and actions

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

-0.18

occan

-0.15

usters

-0.15

IU

-0.14

_backup

-0.14

usted

-0.14

Magnitude

-0.14

itespace

-0.14

het

-0.13

POSITIVE LOGITS

liga

0.16

ERING

0.15

ments

0.15

Ú¯Ø±

0.15

mt

0.15

_mE

0.14

inho

0.14

esel

0.14

ment

0.14

VML

0.14

Activations Density 0.009%

terms related to dispatching or sending out responses and actions

No Comments

No Known Activations