INDEX

Explanations

mentions of the role or function of various entities in a given context

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "role" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

blast

-0.83

iken

-0.75

trap

-0.72

burst

-0.71

slaught

-0.68

itialized

-0.66

secure

-0.66

launch

-0.63

nings

-0.63

[|

-0.62

POSITIVE LOGITS

 intermedi

0.83

 informants

0.81

 stakeholders

0.77

 intermediary

0.77

 referees

0.72

 women

0.70

 shadow

0.69

 courts

0.69

 elements

0.68

 PACs

0.67

Activations Density 0.096%

mentions of the role or function of various entities in a given context

references to the concept of "role" in various contexts

No Comments

No Known Activations