INDEX

Explanations

words starting with the letter "H"

oai_token-act-pair · gpt-3.5-turbo

the letter "H" in various contexts

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

ãĥ¼ãĥĨãĤ£

-0.80

 Beware

-0.73

enhagen

-0.72

EStream

-0.71

DragonMagazine

-0.70

 Conquest

-0.68

CPR

-0.68

 totality

-0.67

 Clicker

-0.66

 Chocobo

-0.65

POSITIVE LOGITS

ospital

1.12

awk

1.10

IGH

1.09

umble

1.09

anging

1.07

acking

1.07

aven

1.05

ISTORY

1.04

OST

1.04

OLD

1.04

Activations Density 0.048%

words starting with the letter "H"

the letter "H" in various contexts

No Comments

No Known Activations