INDEX

Explanations

key events or actions in a narrative

oai_token-act-pair · gpt-3.5-turbo

significant events or announcements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

taboola

-0.69

 varies

-0.67

continue

-0.67

ocations

-0.65

 continue

-0.64

 Provided

-0.62

anse

-0.61

aspx

-0.60

 continues

-0.60

Drop

-0.59

POSITIVE LOGITS

 himself

0.87

 smack

0.81

 bragging

0.81

 brilliantly

0.80

 triumphant

0.78

 proclaiming

0.78

!.

0.78

 THREE

0.77

 twice

0.77

 midway

0.76

Activations Density 0.779%

key events or actions in a narrative

significant events or announcements

No Comments

No Known Activations