INDEX

Explanations

verbs that indicate an attempt or effort

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

dylib

-0.78

requisite

-0.65

 origin

-0.60

inational

-0.60

usage

-0.59

effect

-0.58

abyte

-0.57

 References

-0.56

VERTISEMENT

-0.56

spread

-0.56

POSITIVE LOGITS

 unsuccessfully

1.08

 desperately

0.97

 harder

0.90

 valiant

0.84

reprene

0.80

 frantically

0.71

 hard

0.70

ãĥīãĥ©ãĤ´ãĥ³

0.68

ioned

0.67

ãĥīãĥ©

0.67

Activations Density 0.061%

verbs that indicate an attempt or effort

No Comments

No Known Activations