INDEX

Explanations

references to positive attributes or sentiments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

erv

-0.15

áº½

-0.15

endon

-0.15

rer

-0.15

íĬ¼

-0.14

Fog

-0.14

Å¼

-0.14

vey

-0.14

ContentLoaded

-0.14

sav

-0.13

POSITIVE LOGITS

-negative

0.27

/n

0.23

IntegerField

0.19

/-

0.18

itives

0.18

-sum

0.17

-feedback

0.17

_integer

0.17

_INFINITY

0.17

cly

0.16

Activations Density 0.023%

references to positive attributes or sentiments

No Comments

No Known Activations