INDEX

Explanations

references to the finale of a TV series or season

oai_token-act-pair · gpt-3.5-turbo

words related to "finales" and "climaxes" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

olds

-0.75

guns

-0.74

tics

-0.73

gun

-0.73

ez

-0.73

lying

-0.72

var

-0.70

articles

-0.68

holding

-0.68

held

-0.68

POSITIVE LOGITS

 finale

1.13

adelphia

0.92

 prematurely

0.87

 reel

0.81

 adolesc

0.80

 spoiler

0.79

 showdown

0.79

[|

0.77

 Recap

0.74

APR

0.73

Activations Density 0.011%

references to the finale of a TV series or season

words related to "finales" and "climaxes" in various contexts

No Comments

No Known Activations