INDEX

Explanations

adjectives expressing extreme positivity or negativity

oai_token-act-pair · gpt-3.5-turbo

terms related to scale, particularly large or significant concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tops

-0.77

THER

-0.64

GH

-0.63

ãĥĹ

-0.63

PLA

-0.62

ighth

-0.61

ãĥīãĥ©ãĤ´ãĥ³

-0.61

ĳå£«

-0.61

ãĥĻ

-0.61

asio

-0.60

POSITIVE LOGITS

too

0.73

 lest

0.63

iated

0.62

 bait

0.62

 cooks

0.61

 lately

0.60

aban

0.59

 compliment

0.58

amus

0.57

sighted

0.57

Activations Density 0.105%

adjectives expressing extreme positivity or negativity

terms related to scale, particularly large or significant concepts

No Comments

No Known Activations