INDEX

Explanations

pronouns followed by possessive adjectives

oai_token-act-pair · gpt-3.5-turbo

references to possession or ownership

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

Gallery

-0.77

 Unsure

-0.71

stad

-0.71

tu

-0.69

 Hastings

-0.67

rette

-0.67

AAAA

-0.66

FP

-0.66

pc

-0.66

hoe

-0.66

POSITIVE LOGITS

own

1.44

ELF

1.21

 predecessor

1.03

 respective

1.01

 namesake

0.97

self

0.95

 predecessors

0.94

 newfound

0.88

 rightful

0.88

 entirety

0.86

Activations Density 0.135%

pronouns followed by possessive adjectives

references to possession or ownership

No Comments

No Known Activations