INDEX

Explanations

comparative adjectives ending in -er

oai_token-act-pair · gpt-3.5-turbo

phrases related to increasing tension or complications in various situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

lean

-0.86

leans

-0.79

icle

-0.66

PROV

-0.65

ufact

-0.64

finder

-0.63

fits

-0.63

poon

-0.63

psc

-0.61

Percent

-0.60

POSITIVE LOGITS

 tensions

1.49

 suspicions

1.20

 tension

1.18

 polarization

1.11

 fears

1.10

 mistrust

1.10

 hostilities

1.07

 instability

1.05

 resentment

1.04

 uncertainty

1.04

Activations Density 0.143%

comparative adjectives ending in -er

phrases related to increasing tension or complications in various situations

No Comments

No Known Activations