INDEX

Explanations

phrases related to legitimacy or legal validity

oai_token-act-pair · gpt-3.5-turbo

instances of the word "legitimate" and related concepts

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

hani

-0.92

mith

-0.71

draw

-0.70

azel

-0.70

hire

-0.70

 Leaves

-0.68

hirt

-0.68

emi

-0.67

berries

-0.66

ikers

-0.66

POSITIVE LOGITS

itimate

1.00

 legitimate

0.75

tarian

0.75

 sounding

0.72

acies

0.70

ãĤ®

0.69

 altru

0.68

ulent

0.68

DonaldTrump

0.67

itable

0.66

Activations Density 0.014%

phrases related to legitimacy or legal validity

instances of the word "legitimate" and related concepts

No Comments

No Known Activations