INDEX

Explanations

mentions or references to people in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to people and their opinions or behaviors

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

 srfAttach

-0.77

ãĥ¯

-0.71

éŃĶ

-0.67

paralleled

-0.66

inth

-0.66

yx

-0.64

 Rhodes

-0.64

actory

-0.63

 predecessor

-0.63

orthy

-0.62

POSITIVE LOGITS

 underestimate

0.99

 clam

0.97

 underest

0.96

 misunderstanding

0.91

 misunderstand

0.91

 dying

0.89

 noticing

0.89

 flock

0.88

 afraid

0.87

 hating

0.87

Activations Density 0.231%

mentions or references to people in various contexts

references to people and their opinions or behaviors

No Comments

No Known Activations