INDEX

Explanations

adjectives and nouns related to strong, influential, and dominant entities or personalities

oai_token-act-pair · gpt-3.5-turbo

terms related to influential roles and leadership

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Nightmares

-0.65

ancies

-0.65

 Trails

-0.63

 Goals

-0.62

 Dreams

-0.61

aturdays

-0.59

pleting

-0.58

 Abstract

-0.58

ceivable

-0.58

BW

-0.56

POSITIVE LOGITS

 extraord

1.00

 amongst

0.92

 among

0.85

agonist

0.82

 amidst

0.78

of

0.77

inel

0.76

leader

0.75

 presiding

0.74

 behind

0.74

Activations Density 0.211%

adjectives and nouns related to strong, influential, and dominant entities or personalities

terms related to influential roles and leadership

No Comments

No Known Activations