INDEX

Explanations

instances of the word "check"

oai_token-act-pair · gpt-3.5-turbo

repetitions of the word "check" in various forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 pleas

-0.65

 Archdemon

-0.65

 physi

-0.64

 subject

-0.64

 precip

-0.64

 eminent

-0.63

ram

-0.63

 prolong

-0.62

 frag

-0.62

 retaining

-0.61

POSITIVE LOGITS

check

4.00

checking

2.90

checks

2.77

checked

2.39

Check

1.98

CHECK

1.86

 check

1.62

 Check

1.50

 Checking

1.45

 checking

1.37

Activations Density 0.009%

instances of the word "check"

repetitions of the word "check" in various forms

No Comments

No Known Activations