INDEX

Explanations

proper nouns related to names or titles

oai_token-act-pair · gpt-3.5-turbo

references to the name "Gre" 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£

-0.78

ãĥ¼ãĥ³

-0.78

etheless

-0.74

DragonMagazine

-0.72

CPC

-0.71

ãĤ¢ãĥ«

-0.69

è¦ļéĨĴ

-0.68

ãģ§

-0.67

unct

-0.66

ional

-0.65

POSITIVE LOGITS

yson

0.90

lev

0.87

endale

0.85

ety

0.84

lette

0.84

enery

0.83

edy

0.83

asures

0.82

Gre

0.81

ysc

0.79

Activations Density 0.009%

proper nouns related to names or titles

references to the name "Gre" in various contexts

No Comments

No Known Activations