INDEX

Explanations

words related to human and animal body parts, particularly focusing on heads

oai_token-act-pair · gpt-3.5-turbo

references to heads or head-related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

PsyNetMessage

-0.70

Constructed

-0.68

Suc

-0.67

Ans

-0.67

terday

-0.64

 mell

-0.64

eers

-0.61

 IMAGES

-0.61

 Canad

-0.60

ILLE

-0.60

POSITIVE LOGITS

gear

1.22

dress

1.16

stock

1.13

canon

1.13

scar

1.09

phones

1.09

butt

1.09

liners

1.08

quartered

1.07

quarter

1.05

Activations Density 0.029%

words related to human and animal body parts, particularly focusing on heads

references to heads or head-related concepts

No Comments

No Known Activations