INDEX

Explanations

durations and numeric values associated with media content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

CEF

-0.15

agua

-0.14

elas

-0.14

swire

-0.14

 append

-0.14

INGTON

-0.13

è£

-0.13

opus

-0.13

isan

-0.13

ÄįnÃ½

-0.13

POSITIVE LOGITS

Touches

0.15

ho

0.15

brtc

0.14

ìĪĻ

0.14

ropol

0.13

vt

0.13

arges

0.13

.scalablytyped

0.13

ARG

0.13

parm

0.13

Activations Density 0.059%

durations and numeric values associated with media content

No Comments

No Known Activations