INDEX

Explanations

references to specific individuals involved in reporting or activism

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

ìķ½

-0.15

rox

-0.14

Ã³w

-0.14

Ã³j

-0.14

atak

-0.13

ÑĤÑĮ

-0.13

Å¼e

-0.13

è¡Ĩ

-0.13

isel

-0.13

.Names

-0.13

POSITIVE LOGITS

 Inner

0.32

Inner

0.28

.Inner

0.25

inner

0.24

 inner

0.23

-inner

0.23

 Outer

0.21

 INNER

0.20

.inner

0.20

(inner

0.20

Activations Density 0.000%

references to specific individuals involved in reporting or activism

No Comments

No Known Activations