INDEX

Explanations

phrases related to trust and belief

oai_token-act-pair · gpt-3.5-turbo

expressions of trust and belief in individuals or groups

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

ibaba

-0.79

wich

-0.73

 coma

-0.70

ozo

-0.70

 blur

-0.67

mite

-0.67

ighters

-0.65

TN

-0.65

udi

-0.65

ional

-0.64

POSITIVE LOGITS

worthiness

0.96

enance

0.88

 integrity

0.82

orum

0.79

 trusting

0.77

 sincerity

0.74

isdom

0.74

 confidence

0.71

 assurance

0.71

 legitimacy

0.70

Activations Density 0.172%

phrases related to trust and belief

expressions of trust and belief in individuals or groups

No Comments

No Known Activations