INDEX

Explanations

occurrences of the word "In" in different contexts

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

jected

-0.19

ject

-0.17

tempt

-0.17

cluding

-0.17

 spite

-0.17

/from

-0.16

jective

-0.16

dition

-0.15

gether

-0.15

POSITIVE LOGITS

gers

0.19

ozem

0.19

ez

0.19

cheon

0.19

nes

0.18

uit

0.18

 memor

0.17

Sink

0.17

uits

0.17

Ãªs

0.17

Activations Density 0.041%

occurrences of the word "In" in different contexts

No Comments

No Known Activations