INDEX

Explanations

verbs related to ability or expertise

oai_token-act-pair · gpt-3.5-turbo

indicative verbs and phrases that describe the state of being or existence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

heny

-0.88

 Franch

-0.76

 Awakens

-0.72

ulla

-0.71

orks

-0.71

backer

-0.68

 oneself

-0.67

 ourselves

-0.67

uld

-0.65

 waits

-0.65

POSITIVE LOGITS

 impecc

0.94

 priceless

0.93

 hampered

0.93

 waning

0.87

 questionable

0.86

 mirrored

0.86

 hindered

0.84

 admirable

0.84

 bolstered

0.84

 undeniable

0.83

Activations Density 0.304%

verbs related to ability or expertise

indicative verbs and phrases that describe the state of being or existence

No Comments

No Known Activations