INDEX

Explanations

phrases related to evaluating or critiquing how a situation is managed or dealt with

oai_token-act-pair · gpt-3.5-turbo

references to the management or administration of issues or situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

hemat

-0.75

elong

-0.74

plex

-0.71

uni

-0.69

owicz

-0.64

ode

-0.64

evil

-0.63

root

-0.62

mpeg

-0.62

interrupted

-0.62

POSITIVE LOGITS

bars

0.98

 mishand

0.90

 handled

0.78

ttes

0.78

 handling

0.74

DAQ

0.72

terness

0.72

 guiName

0.70

 handles

0.68

VICE

0.67

Activations Density 0.016%

phrases related to evaluating or critiquing how a situation is managed or dealt with

references to the management or administration of issues or situations

No Comments

No Known Activations