INDEX

Explanations

recently mentioned or previously mentioned items

oai_token-act-pair · gpt-3.5-turbo

references to time indicators related to recent or previously occurred events

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

atism

-0.84

afety

-0.69

ILY

-0.68

ibility

-0.67

 Cohn

-0.67

 DRAG

-0.63

ility

-0.61

olor

-0.60

letico

-0.60

 Integrity

-0.60

POSITIVE LOGITS

existing

1.14

mentioned

1.09

 deceased

1.02

 unseen

1.02

 discussed

0.96

 unsus

0.96

 mentioned

0.95

 released

0.94

 departed

0.92

 occurring

0.91

Activations Density 0.125%

recently mentioned or previously mentioned items

references to time indicators related to recent or previously occurred events

No Comments

No Known Activations