INDEX

Explanations

words ending with '-ern'

oai_token-act-pair · gpt-3.5-turbo

words related to "ern," possibly common suffixes or themes associated with characters or settings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 collagen

-0.65

xon

-0.65

 concluding

-0.65

tin

-0.64

 inhibitors

-0.63

 trickle

-0.62

 retreating

-0.61

 Moderate

-0.61

 boiled

-0.61

 priced

-0.60

POSITIVE LOGITS

opol

1.13

acle

1.04

ern

1.01

acular

0.97

esse

0.94

anke

0.91

ophon

0.90

ophe

0.90

atural

0.89

acles

0.89

Activations Density 0.011%

words ending with '-ern'

words related to "ern," possibly common suffixes or themes associated with characters or settings

No Comments

No Known Activations