INDEX

Explanations

noun phrases indicating different types or examples of something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "as"

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

raq

-0.74

ranean

-0.72

oes

-0.70

OSE

-0.70

ATIONAL

-0.68

rax

-0.68

YD

-0.67

itiveness

-0.66

BACK

-0.65

OUS

-0.65

POSITIVE LOGITS

pired

0.96

lihood

0.91

 ours

0.88

pects

0.85

 well

0.74

pires

0.71

 regards

0.71

 evidenced

0.70

etheless

0.69

pire

0.68

Activations Density 0.108%

noun phrases indicating different types or examples of something

instances of the word "as"

No Comments

No Known Activations