INDEX

Explanations

references to specific individuals' names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

gratis

-0.16

lops

-0.15

[$_

-0.15

ĴĮ

-0.15

 ,č↵

-0.15

llen

-0.14

 createState

-0.14

lub

-0.14

æĸ¯çī¹

-0.14

Terminal

-0.14

POSITIVE LOGITS

 trace

0.19

 Trace

0.17

trace

0.17

 below

0.17

_trace

0.16

below

0.16

 traces

0.16

(trace

0.16

0.15

 Madd

0.15

Activations Density 0.021%

references to specific individuals' names

No Comments

No Known Activations