INDEX

Explanations

adjectives or verb phrases indicating judgments or opinions

oai_token-act-pair · gpt-3.5-turbo

words and phrases related to existence, reality, and belief assertions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

çīĪ

-0.66

rises

-0.64

aughters

-0.64

ofi

-0.63

 winds

-0.62

imar

-0.62

eatures

-0.62

ispers

-0.58

 refres

-0.58

cellaneous

-0.58

POSITIVE LOGITS

?,

1.01

izable

0.94

 anymore

0.87

 somehow

0.82

acea

0.79

enough

0.79

/,

0.79

 someday

0.77

achable

0.73

.*

0.72

Activations Density 0.627%

adjectives or verb phrases indicating judgments or opinions

words and phrases related to existence, reality, and belief assertions

No Comments

No Known Activations