INDEX

Explanations

phrases indicating someone taking on a specific role, responsibility, or challenge

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "take on."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ONSORED

-0.74

States

-0.70

 specificity

-0.68

WA

-0.68

Anim

-0.66

 Bullets

-0.65

oiler

-0.65

Compar

-0.65

Loading

-0.64

Po

-0.64

POSITIVE LOGITS

erous

1.07

emonic

0.85

shore

0.84

unal

0.82

etime

0.79

 behalf

0.78

*/(

0.78

 board

0.77

Downloadha

0.76

 occasion

0.75

Activations Density 0.035%

phrases indicating someone taking on a specific role, responsibility, or challenge

instances of the phrase "take on."

No Comments

No Known Activations