INDEX

Explanations

words related to large, powerful, and dominant entities or forces

oai_token-act-pair · gpt-3.5-turbo

references to juggling or juggernauts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 sentencing

-0.68

 condem

-0.66

 hospitality

-0.65

wards

-0.65

 tasting

-0.65

 judgement

-0.62

ibaba

-0.61

Interstitial

-0.61

 unconditional

-0.61

raviolet

-0.61

POSITIVE LOGITS

ernaut

1.49

 jugg

1.10

glers

1.05

 juggling

0.98

alo

0.96

alos

0.91

adelphia

0.91

annon

0.88

arro

0.84

ãĥ¼ãĤ¯

0.83

Activations Density 0.019%

words related to large, powerful, and dominant entities or forces

references to juggling or juggernauts

No Comments

No Known Activations