INDEX

Explanations

phrases related to probability and likelihood

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate probabilities and chances of events occurring

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 ILCS

-0.92

minus

-0.84

ngth

-0.82

umbered

-0.79

zek

-0.79

geist

-0.77

umbers

-0.75

culated

-0.74

ahime

-0.74

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.73

POSITIVE LOGITS

 encountering

1.41

 obtaining

1.36

 succeeding

1.35

 getting

1.29

 success

1.28

 achieving

1.28

 winning

1.27

 finding

1.23

 catching

1.23

 acquiring

1.20

Activations Density 0.150%

phrases related to probability and likelihood

phrases that indicate probabilities and chances of events occurring

No Comments

No Known Activations