INDEX

Explanations

phrases related to problem-solving or solution-finding

oai_token-act-pair · gpt-3.5-turbo

phrases indicating methods or paths to achieve goals or solutions

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

inent

-0.77

eatures

-0.74

ewitness

-0.72

hovah

-0.69

amaz

-0.69

asts

-0.69

inently

-0.67

 livest

-0.64

usters

-0.64

uster

-0.63

POSITIVE LOGITS

forward

0.88

 somew

0.83

fare

0.82

 workaround

0.80

 circumvent

0.76

finding

0.74

 forward

0.73

point

0.70

 backdoor

0.69

 whereby

0.69

Activations Density 0.061%

phrases related to problem-solving or solution-finding

phrases indicating methods or paths to achieve goals or solutions

No Comments

No Known Activations