INDEX

Explanations

words related to neutrality or lack of bias

oai_token-act-pair · gpt-3.5-turbo

content related to neutrality and its various implications

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

KER

-0.73

monary

-0.72

hower

-0.71

teenth

-0.66

teen

-0.64

inary

-0.64

Requires

-0.62

buck

-0.62

 painfully

-0.62

urrent

-0.62

POSITIVE LOGITS

 confines

0.90

 zone

0.87

 toward

0.83

 stance

0.80

 reception

0.79

 towards

0.79

glers

0.77

 environment

0.77

 demeanor

0.74

 matchups

0.71

Activations Density 0.078%

words related to neutrality or lack of bias

content related to neutrality and its various implications

No Comments

No Known Activations