INDEX

Explanations

phrases related to specific actions or instructions

oai_token-act-pair · gpt-3.5-turbo

references to personal experiences or identifiable individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Casey

-0.98

 Amon

-0.82

 Kats

-0.82

 Mish

-0.82

 Misty

-0.80

AH

-0.79

 Leah

-0.79

 Slay

-0.77

Sah

-0.77

POSITIVE LOGITS

 Rodrig

0.96

PORT

0.93

Wilson

0.91

 Winston

0.91

Pen

0.91

 Rodgers

0.91

Pen

0.90

inson

0.90

 Wilson

0.88

pen

0.87

Activations Density 0.353%

phrases related to specific actions or instructions

references to personal experiences or identifiable individuals

No Comments

No Known Activations