INDEX

Explanations

verbs and adjectives related to assessing and evaluating

oai_token-act-pair · gpt-3.5-turbo

terms related to evaluation and assessment metrics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

dt

-0.72

quished

-0.72

£ı

-0.72

dding

-0.70

rt

-0.67

ffe

-0.66

 Originally

-0.65

entimes

-0.65

elected

-0.64

lyn

-0.64

POSITIVE LOGITS

 effectiveness

1.81

 adequ

1.61

 efficacy

1.55

 viability

1.53

 feasibility

1.52

worthiness

1.48

 usefulness

1.40

 merits

1.39

 readiness

1.38

 competence

1.35

Activations Density 0.310%

verbs and adjectives related to assessing and evaluating

terms related to evaluation and assessment metrics

No Comments

No Known Activations