INDEX

Explanations

phrases related to actions or occurrences that involve a strong level of involvement or influence by individuals or groups

oai_token-act-pair · gpt-3.5-turbo

verbs related to claims, accusations, and interventions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

gae

-0.79

enei

-0.75

alg

-0.75

 resultant

-0.66

 attest

-0.64

headers

-0.63

iasm

-0.63

gel

-0.61

ofi

-0.60

 healed

-0.59

POSITIVE LOGITS

 inappropriately

0.84

 wrong

0.68

 improperly

0.67

 wrongly

0.66

 Paddock

0.66

netflix

0.63

 illegally

0.63

"))

0.62

 beforehand

0.61

 unfairly

0.61

Activations Density 0.419%

phrases related to actions or occurrences that involve a strong level of involvement or influence by individuals or groups

verbs related to claims, accusations, and interventions

No Comments

No Known Activations