INDEX

Explanations

phrases related to quotes or reported speech

oai_token-act-pair · gpt-3.5-turbo

instances of quoted speech or dialogue

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

folios

-0.73

cffffcc

-0.73

swick

-0.67

ãĥ¼ãĥĨ

-0.66

aired

-0.66

nerg

-0.66

 overfl

-0.65

 blows

-0.64

ptives

-0.63

Es

-0.62

POSITIVE LOGITS

omething

1.06

ynthesis

0.93

ometimes

0.92

creen

0.91

paces

0.89

ysis

0.87

ynt

0.86

 goodbye

0.84

pace

0.81

cale

0.78

Activations Density 0.083%

phrases related to quotes or reported speech

instances of quoted speech or dialogue

No Comments

No Known Activations