INDEX

Explanations

adjectives or phrases indicating difficulty or negativity

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate difficulty or negative conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Icar

-0.72

ãĥĨ

-0.72

pert

-0.68

last

-0.66

Offline

-0.66

isol

-0.64

tie

-0.62

 Kinder

-0.60

thro

-0.60

urd

-0.59

POSITIVE LOGITS

agher

0.68

uce

0.67

 Carbuncle

0.65

oshenko

0.64

 suffice

0.64

heny

0.61

 antioxidants

0.61

quartered

0.59

 entr

0.59

è¦ļéĨĴ

0.59

Activations Density 0.215%

adjectives or phrases indicating difficulty or negativity

phrases that indicate difficulty or negative conditions

No Comments

No Known Activations