INDEX

Explanations

adjustment-related terms or actions

oai_token-act-pair · gpt-3.5-turbo

terms related to adjustment and adaptation

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

gom

-0.79

OGR

-0.67

chu

-0.66

 Chrys

-0.66

advertisement

-0.65

hex

-0.64

hov

-0.63

ijk

-0.63

intel

-0.62

join

-0.62

POSITIVE LOGITS

 adjustments

0.96

 adjustment

0.93

 adjusting

0.92

 adjust

0.90

ments

0.83

 adjusts

0.81

imar

0.79

 adjusted

0.77

 balances

0.76

 Adjust

0.76

Activations Density 0.043%

adjustment-related terms or actions

terms related to adjustment and adaptation

No Comments

No Known Activations