INDEX

Explanations

mentions of the word "af"

oai_token-act-pair · gpt-3.5-turbo

repeated mentions of the suffix "af"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Olymp

-0.67

Dru

-0.67

 fixation

-0.66

 Boone

-0.65

 Hulk

-0.65

 Wink

-0.64

MLG

-0.63

 Creator

-0.63

 Patriarch

-0.62

 paraly

-0.60

POSITIVE LOGITS

rican

1.41

rica

1.30

ayette

1.21

avorite

1.11

amily

1.10

rost

1.06

ornia

1.04

eatures

1.04

raid

1.04

onso

1.04

Activations Density 0.016%

mentions of the word "af"

repeated mentions of the suffix "af"

No Comments

No Known Activations