INDEX

Explanations

phrases related to TV shows or programs

oai_token-act-pair · gpt-3.5-turbo

mentions of "show," indicating a focus on television or performance-related content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 neutral

-0.74

 goals

-0.66

 membranes

-0.65

 liter

-0.65

 leveled

-0.61

etooth

-0.61

arily

-0.60

 ends

-0.60

dec

-0.60

 airborne

-0.60

POSITIVE LOGITS

show

4.11

shows

2.86

Show

2.18

 Show

1.89

 show

1.86

 SHOW

1.78

 Shows

1.57

display

1.35

episode

1.28

hide

1.26

Activations Density 0.008%

phrases related to TV shows or programs

mentions of "show," indicating a focus on television or performance-related content

No Comments

No Known Activations