INDEX

Explanations

pronouns referring to people or entities previously mentioned in the text

oai_token-act-pair · gpt-3.5-turbo

references to people or groups in relation to their actions or characteristics

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

AND

-0.71

onsequ

-0.64

 Meaning

-0.62

ising

-0.62

PLIC

-0.60

 Trop

-0.59

ont

-0.58

ounding

-0.57

istance

-0.56

Instruct

-0.56

POSITIVE LOGITS

 owns

1.10

 specializes

1.00

 specialize

0.95

 participated

0.92

 loves

0.88

 understands

0.87

 possesses

0.86

 enjoys

0.85

 cares

0.84

 participates

0.84

Activations Density 0.118%

pronouns referring to people or entities previously mentioned in the text

references to people or groups in relation to their actions or characteristics

No Comments

No Known Activations