INDEX

Explanations

statements indicating potential consequences or implications

oai_token-act-pair · gpt-3.5-turbo

references to costs associated with projects or actions

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

rontal

-0.70

ukong

-0.69

kay

-0.67

acca

-0.66

ribune

-0.66

rill

-0.65

izza

-0.65

letes

-0.65

cision

-0.64

uminati

-0.64

POSITIVE LOGITS

 lacked

1.17

 lacks

1.06

 cautioned

1.04

 hindered

1.00

 overshadowed

0.96

 differed

0.93

 stalled

0.93

 hampered

0.92

 balk

0.92

 fails

0.91

Activations Density 0.749%

statements indicating potential consequences or implications

references to costs associated with projects or actions

No Comments

No Known Activations