INDEX

Explanations

phrases related to power dynamics and examples of items representing different concepts

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate examples or references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

etheless

-0.79

ãĤ©

-0.76

è¦ļéĨĴ

-0.69

livion

-0.65

ateur

-0.63

chant

-0.61

jug

-0.61

Ŀ

-0.60

unda

-0.59

 sidx

-0.59

POSITIVE LOGITS

 example

1.32

for

1.31

 instance

1.27

for

1.10

say

1.10

For

0.97

example

0.91

say

0.89

For

0.85

 examples

0.78

Activations Density 0.427%

phrases related to power dynamics and examples of items representing different concepts

phrases that indicate examples or references

No Comments

No Known Activations