INDEX

Explanations

proper nouns related to individuals or entities

oai_token-act-pair · gpt-3.5-turbo

key names and significant numerical data related to political contexts and events

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

`.

-0.77

================================

-0.70

%).

-0.69

 attRot

-0.68

âķĲâķĲ

-0.68

ÂłÂłÂł

-0.67

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

-0.67

''.

-0.66

"""

-0.65

-0.64

POSITIVE LOGITS

 twice

0.96

 while

0.91

 alongside

0.91

 before

0.84

 amid

0.84

 whereas

0.83

 during

0.83

but

0.81

 afterward

0.79

 when

0.76

Activations Density 0.931%

proper nouns related to individuals or entities

key names and significant numerical data related to political contexts and events

No Comments

No Known Activations