INDEX

Explanations

words related to exaggeration or intensity

oai_token-act-pair · gpt-3.5-turbo

terms associated with conditions of excess or extreme states

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

emouth

-0.70

baum

-0.68

ABE

-0.67

iage

-0.65

çĦ

-0.63

SPONSORED

-0.61

wick

-0.61

 Grounds

-0.60

 Folder

-0.60

ulhu

-0.59

POSITIVE LOGITS

ulic

0.90

emic

0.84

uating

0.80

opter

0.80

catentry

0.77

ilic

0.76

ersive

0.73

inem

0.73

nir

0.71

etch

0.68

Activations Density 0.111%

words related to exaggeration or intensity

terms associated with conditions of excess or extreme states

No Comments

No Known Activations