INDEX

Explanations

descriptions or mentions of interviews

oai_token-act-pair · gpt-3.5-turbo

occurrences of interviews or interview-related contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

vir

-0.82

_{

-0.79

 upstream

-0.74

 downstream

-0.73

 launchers

-0.69

 Implementation

-0.65

åĮ

-0.65

Wad

-0.65

 implementations

-0.64

 levy

-0.64

POSITIVE LOGITS

 interview

1.66

 interviews

1.45

 interviewer

1.39

Interview

1.39

 Interview

1.38

 candid

1.31

 taped

1.26

 podcast

1.24

 Vanity

1.17

 interviewed

1.14

Activations Density 0.361%

descriptions or mentions of interviews

occurrences of interviews or interview-related contexts

No Comments

No Known Activations