INDEX

Explanations

instances where specific criteria are met or where actions are required

oai_token-act-pair · gpt-3.5-turbo

terms related to legal or regulatory requirements for license or approval processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

tc

-0.69

ãĥ³

-0.67

"]=>

-0.64

roo

-0.64

RIP

-0.64

 underest

-0.63

NetMessage

-0.63

****************

-0.63

????

-0.62

 metaphors

-0.61

POSITIVE LOGITS

 satisfactory

1.11

 satisfied

0.99

 meets

0.98

 demonstrate

0.97

atisf

0.96

 fulfilled

0.95

 satisfies

0.94

 certify

0.93

 qualifying

0.93

 completed

0.91

Activations Density 0.357%

instances where specific criteria are met or where actions are required

terms related to legal or regulatory requirements for license or approval processes

No Comments

No Known Activations