INDEX

Explanations

occurrences of the term "neutral" and its related forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

onto

-0.15

elson

-0.15

ependency

-0.15

#ad

-0.14

elerik

-0.14

away

-0.14

elman

-0.14

 Hulu

-0.14

lah

-0.14

.wr

-0.14

POSITIVE LOGITS

ized

0.20

ised

0.20

ize

0.18

ÑĢÐ¾Ð³

0.17

izing

0.17

 third

0.16

 Lloyd

0.16

 neutral

0.15

na

0.15

ana

0.15

Activations Density 0.010%

occurrences of the term "neutral" and its related forms

No Comments

No Known Activations